McMillan 28
officials as well as archives of each state’s major newspapers. With little institutional memory in
governor’s offices and turnover in economic development agencies, it is difficult to find officials
who remember governor-led missions each year. Although it is possible for newspaper searches
to have missed some missions, archives seem to be a reliable source of data.
C
ONTROL
V
ARIABLES
Several state economic and educational variables are used as control variables for this
analysis. It is important to include socioeconomic variables to show differences between states
because overseas firms are likely to take educational training, employment opportunities
available, and the cost of doing business into account when selecting a state in which to locate.
States’ unemployment rates from the Bureau of Labor Statistics (2006) are used and state and
local tax burdens for each year are incorporated from The Tax Foundation (2006)
since state
officials say financial incentives can be important to investment attraction (Personal Interviews
2006). High school dropout rates (1993-2002) are used from a U.S. Department of Education
report by Tab (2006), employed as a lagged variable in each model such that the dropout rate for
1993-94 is reported for 1995. Lagging this variable is theoretically beneficial because rates are
generally not reported until the following year, when they might begin to affect investment
decisions. This method also yields more data points since several states do not subscribe to the
federal formula for dropout rates, leaving only 165 observations.
A
NALYSIS
,
R
ESULTS AND
I
NTERPRETATION
Four different models are tested using TSCS analysis in a fixed effects model. It is
theoretically important to lag both the number of governor-led trade missions and states’
international offices to see if they have an effect upon a state’s investment attraction efforts in
19
States’ corporate income tax rate was also collected from The Tax Foundation, but had too much missing data for
the years of this study.
Convention