Page 100 - 33-3

Goal Consensus, Subordinates’ Prior Performances, and Supervisors’ Resource Allocation Preferences

discretionary allocations; nevertheless, branch-specific characteristics (e.g., the larger the
size of the branch office, the greater the need for sales personnel and advertising funding),
the seasonality of the housing market, and the regional manager’s preferences toward
each branch office all affect the regional manager’s resource allocation decision on these

offices. In other words, both the resources that a branch office requires to maintain its daily
business as well as the regional manager’s personal “preferences” toward the branch office
influence the regional manager’s resource allocation decision.

To measure the “unexpected” portion of resources in each branch office, we regress
each branch office’s monthly ratio of the number of senior sales agents to the total number
7
of sales agents (SENIOR ) and the allocated advertising funding (ADV ) on the following
i,t
i,t
key branch-specific characteristic variables: branch size (OFFICE_SIZE ), branch
i,t
age (OFFICE_AGE ), branch manager’s service period (MGT_TENURE ), targeted
i,t
i,t
sales revenue of the branch office (TARGET ), average property price per deal sold by
i,t
the branch office (PRICE ), number of houses managed by the branch office (NUM_
i,t
ENTRUST ), and total number of branch offices in the same region (NUM_BRANCH ).
i,t
i,t
We also include months and regions to control for time and regional effects that could be
conjunct with the regional manager’s resource allocation decision. The residuals from
these two models represent the unexpected portion (preferences) of resources distributed
to the branch office. A higher value residual term indicates that more unexpected resources
are given to the branch office, and a lower residual term indicates that fewer unexpected

resources are allocated to the branch office. The regression results of equations (1) and (2)
are shown in Appendix C, and detailed variable definitions are presented in Table 2. The
regression models are expressed as follows:

SENIOR i,t = α +α OFFICE_SIZE + α MGT_TENURE + α TARGET + α PRICE i,t
2
i,t
i,t
4
i,t
3
1
0
+ α NUM_ENTRUST + α NUM_BRANCH + ∑MONTH + ∑REGION
5
i,t
i,t
6
+ δ , (1)
i,t
7 Senior salespersons are valuable human resources for branch offices. Senior salespersons are more
capable of handling complex job-related circumstances and sharing organizational knowledge, and
hence contribute toward improving organizational performance in the long run, as the employee
service period length reflects an upward-sloping tenure-productivity profile (McDaniel, Schmidt, and
Hunter, 1988; Wright and Bonett, 2002; Abraham and Medoff, 1985).
92