Modeling industrial location decisions in U.S. counties

all parameters of interest. To see this, let

_   exp(θ⁰ytj + βztjk + γ j )

^Pt/^kk   PJ=1 ^eχP(^θ0 ytj + ^β0ztjk + Yj )

be the probability that the investor at time t selects location j , conditional
on his choice of sector k. Proceeding in a similar fashion as above we can
”condition-out” the local fixed effects and obtain estimates for the β and θ
vectors.¹³

4 An Empirical Application: Locational Determi-
nants of Manufacturing Plant Births Across the
U.S. Counties

4.1 Data and variables

To demonstrate ways to exploit the Poisson-CLM relation as described in
the last section, we give an illustration of firm location decisions where there
are many spatial choices. Specifically, we model the location determinants
of manufacturing plant births for the 3,066 counties belonging to the 48
contiguous U.S. states¹⁴. To take advantage of the relation between the
CLM and Poisson regression, the dependent variable formed for the tests
is the number of establishment births for each county by industry (2-digit
SIC code for all establishments in the manufacturing sector). We use special
U.S. Census Bureau tabulations of the Standard Statistical Establishments
List encompassing the universe of all new known openings for the years of
1989 and 1997. In Tables 1 and 2 we show the industry sector and spatial

13

<<   12   13   14   15   16   17   18   >>

More intriguing information