Page 113 - 33-3

NTU Management Review Vol. 33 No. 3 Dec. 2023

and re-examine whether our results remain for this sub-sample. Branch-monthly
observation is excluded if the office’s senior salespersons ratio (allocating advertising
funding) in the (t-1)-th month equals the senior salespersons ratio (allocating advertising
funding) in the t-th month.

The empirical results are presented in Table 7. From columns (1) and (2) in Table
7, Ab_ADV is positively correlated with CONSENSUS (t = 1.93, p < 0.1) but not Ab_
i,t
i,t
SENIOR , which is consistent with what we find in Table 5. Columns (3) and (4) in Table
i,t
7 are similar to those in Table 5. That is, subordinates’ prior performance (PERF i, t-1 ) is
positively related to Ab_SENIOR (t = 1.80, p < 0.1) but not Ab_ADV . Finally, columns
i,t
i,t
(5) and (6) in Table 7 show that the product term of goal consensus and the branch offices’
past performance is statistically insignificant in relation to either Ab_SENIOR or Ab_
i,t
ADV .
i,t

4.3.3 Effect of Goal Consensus on Relationship between Resource Allocation Preferences
and Abnormal Sales Performance

In addition to investigating the association among goal consensus, the branch
manager’s prior performance, and the regional manager’s resource allocation preferences,
we further examine whether the regional manager’s resource allocation preferences lead to
superior sales performance for the branch in the future. In addition, we investigate whether
goal consensus reinforces their association.

To measure the abnormal sales performance of a branch office, we regress the office’s
total sales revenue on the key branch-specific characteristic variables. The higher the
10
residual term, indicating that the office has more unexpected sales performance, and

vice versa. We then regress the office’s abnormal sales performance at one period lag of
abnormal resource allocation with a set of control variables that could influence branch
performance (Ghosh and Lusch, 2000; Fu, 2009; Lee, Sheldon, and Turban, 2003; Hansen
and Wernerfelt, 1989).

10 We include number of houses managed, average property price per deal sold by branch office, branch
office’s age, office manager’s tenure, and number of branch offices into the regression model, in
addition to month and region fixed effects, as these variables could affect the sales performance of a
branch office (Ghosh and Lusch, 2000; Fu, 2009; Lee et al., 2003).

105