accounting earnings we found model coefficient for the interaction term was

# Accounting earnings we found model coefficient for

This preview shows page 12 - 14 out of 22 pages.

accounting earnings, we found (model coefficient for the interaction term was significant (p < .05), albeit with only a tribution to explanatory power. To bet stand this relationship, we graphed the of financial alignment with firm-specif centered values one standard deviation below the mean, holding everything els Figure 1 presents these slopes. Financia yields a greater payoff to firms when specific wealth is high (? = 120.37.) than low (? = -56.03). This result is consi Hypothesis 4. Nevertheless, despite the significance of the interaction, the small 7 To check for the stability of this estimate, in results not shown here, we reestimated the model without the controls and, alternatively, by including volatility and CEO tenure as additional controls. The results are robust (stable coefficient magnitude and a 95% confidence in terval that, though broad, excludes zero), irrespective of the control variables placed in the model. This content downloaded from 129.125.19.61 on Thu, 07 Feb 2019 14:56:07 UTC All use subject to
1040 Academy of Management Journal October TABLE 3 Results of Fixed-Effects Regression Analysis Predicting CEO Returna CEO Returnb Predictors Model 1 Model 2 Intercept Year dummies 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Controls Volatility Assets Industry return ROA one-yearc CEO tenure Shareholder return0 R2 (dummy variable approach)*1 Adjusted R2 (dummy variable approach)0 Incremental adjusted R2 Adjusted R2 (within)d F 15.48*** (1.05) 5.65*** (0.70) -2.08*** -2.38*** -2.05*** -1.73*** -1.47*** -2.61*** -1.99*** -1.26*** -0.92*** -1.50*** 0.26 (0.26) (0.27) (0.27) (0.25) (0.23) (0.21) (0.19) (0.17) (0.17) (0.18) (0.18) -2.26*** (0.53) -1.40*** (0.12) 0.04*** (0.00) 0.02*** (0.00) -0.05*** (0.01) .40 .22 .19 112.41*** -0.74*** -0.08 -0.88*** -0.59*** -0.72*** -0.66*** -0.13 -0.24* -0.12 0.05 0.04 (0.17) (0.18) (0.18) (0.17) (0.15) (0.14) (0.12) (0.12) (0.12) (0.12) (0.11) -0.01 (0.35) -0.38*** (0.08) 0.00 (0.00) -0.05* (0.02) -0.05*** (0.01) 0.80* .75 .67 .45*** .66 800.49*** (0.01) a Standard errors are in parentheses, n(observations) = 9,168; n(companies) = 2,166. The 95% confidence interval for share holder return is .78-.81. Models using three- and five-year CEO and shareholder returns were also tested, with the results sub stantively confirming the findings presented in this table. These results are available from the authors upon request. b Transformed using the inverse hyperbolic sine function (IHS). c The ROA one-year value was divided by 10 for interpretation. d Primary R2 and adjusted R2 values are reported for this model using dummy variable regression (STATA's "areg" func tion) to estimate fixed effects (coefficients for firm-specific dummy variables have been omitted for brevity). For informa tion purposes, we have also included the adjusted model R2 (within) from the alternative, time-demeaning approach (STATA's "xtreg, fe" function?"fixed-effects" or "within" transformation). Parameter estimates, standard errors, etc., are the same under either approach (Woodridge, 2002, 2009). * < .05 *** < .001 Two-tailed tests. to variance explained is perhaps at least as note worthy. Together with the fact that the interaction between a CEO's total firm-specific wealth and fi nancial alignment does not appear to impact future shareholder return, support for Hypothesis 4 is, all told, somewhat limited.

#### You've reached the end of your free preview.

Want to read all 22 pages?

• Fall '14
• Killian