assumption of independent errors is an important one, because, if violated, it
can lead to biased coefficient estimates. In practice, as shown in this paper,
the empirical studies of location have been unable to fully accommodate the
IIA problem within the CLM. Also, the proposed solutions to accommodate
complex choice scenarios with the decision maker confronting many (nar-
rowly defined) spatial alternatives have been unsatisfactory. More recent
studies on industrial location have tackled this later problem by applying
Poisson (count) models. Yet this direction in empirical modeling has not
been cast as part of the Random Utility Maximization framework, a main
advantage of the McFadden-Carlton approach since it links empirical work
to theory.
Here we show how one can more effectively control for the potential
IIA violation in complex choice scenarios, regardless of the spatial choice
set dimension. This is done by taking advantage of an equivalence rela-
tion between the likelihood functions of the conditional logit model and the
Poisson regression (Guimaraes, Figueiredo & Woodward 2002). We also
provide an empirical illustration, wherein we demonstrate how that relation
can be helpful to provide more reliable estimates for the location determi-
nants of start-up manufacturing plants in the United States counties. We
find strong evidence that agglomeration economies (both urbanization and
localization), as well as taxes, influence location decisions. These relations
hold across all tested specifications, even when we add stringent controls
to account for omitted relevant variables. The evidence concerning other
factors (labor costs, land costs, and local markets) is not as conclusive.
The rest of the paper is comprised of four sections. The next section