Executive Compensation; Regression Analysis A recent study of the airline industry examined whether a performance for a selected nonfinancial measure was a significant predictor of CEO compensation. A sample of 35 firms was taken and regression obtained to determine the potential relationship between selected financial and nonfinancial independent variables and three dependent variables: (a) CEO cash compensation, (b) CEO compensation in the form of options granted during the year, and (c) total CEO compensation (a + b).
The independent variables were:
• Passenger load (PL), the proportion of seats filled to the seats available.
• Stock price return (RT), the increase in stock price plus dividends over the year relative to begin- ning of year.
• Return on assets (ROA), income over total assets.
• Sales.
• Stock price volatility (V), the standard deviation of daily stock price changes in the company's stock price.
• CEO ownership (CEO), the percentage of the company's outstanding shares owned by the CEO.
• CEO tenure(CEO-T), the number of years the CEO has been on the job.
• Ratio of book value of the company to the market value of the company (BM), a measure of the market value of the company.
The table below shows the three dependent variables, the eight independent variables, and the sig- nificance (p-value) of the independent variable in each equation. For example, in the regression with the dependent variable, cash compensation (Regression One), the PL variable is significant at the .01 level, RT is significant at the .01 level, and the ROA variable is not significant. The authors of the study hypothesized that there would be a positive significant relationship between CEO compen- sation and the one nonfinancial variable, passenger load.
| |
Regression One* |
Regression Two* |
Regression Three*
|
|
Passenger load (PL)
|
0.01
|
NS
|
NS |
|
Stock return (RT)*
|
0.01
|
0.01
|
0.01
|
|
Return on assets (ROA)
|
NS
|
NS
|
NS
|
|
Sales*
|
0.01
|
0.01
|
0.01
|
|
Stock price volatility (V)
|
NS
|
NS
|
0.05
|
|
CEO ownership % (CEO)†
|
0.01
|
0.01
|
0.01
|
|
CEO tenure (CEO-T)
|
0.01
|
NS
|
0.05
|
|
Book value to Market Value (BM)
|
NS
|
0.05
|
NS
|
|
R squared
|
69.3%
|
11.0%
|
50.0%
|
* Dependent variables were transformed using the natural logarithm.
† This variable had a significant negative coefficient, indicating an inverse relationship with the dependent variable.
Required Review the three regressions above and develop a brief explanation for each of the following:
1. Which of the three regressions would you most rely on, and why?
2. What do the regression results tell you about the relationships of the independent variables to the three dependent variables?
3. Were the authors of the study correct about their expectation regarding the PL variable?
4. How would you use this information in designing compensation plans for executives in the airline industry?