the model demonstrates that imitation is consistent with expected profit maximiza-
tion, although it usually results in lower aggregate profit than would be achieved by
a benevolent central planner. The benchmark relative to which efficiency is measured
does not assume a new institution for pooling information resources. Rather, it takes
the sequence of moves as given and simultaneously chooses all firms’ choice variables
(i.e., Firm 1 and Firm 2’s quantities of private information acquisition and choices of
location) to maximize aggregate profit. Firms are assumed to be identical except that
later movers can make use of earlier movers’ locations as an additional conditioning
variable.
As emphasized already, imitation in location choice is not uniformly bad for ag-
gregate efficiency. There is a genuine positive externality flowing from early to later
movers. Imitation usually helps exploit this positive externality to some extent, but
not as fully as aggregate efficiency requires. As shown in Figure 2, over most of the
range of the cost-of-information parameter c, the central planner prescribes more in-
formation for Firm 1 and less for Firm 2. Thus, in this parameter range, the central
planner fully exploits the positive externality by increasing, not reducing, the extent
of imitation. When c is very large, however, the central planner requires that both
firms acquire additional private information, implying that a reduction in imitation
is needed to achieve social efficiency. This case might argue for public provision of
neighborhood-level demographic and crime information that can be used to estimate
revenues and costs, or perhaps direct subsidies for first movers into neighborhoods
seeking (re-)development.
Thus, whether imitation is consistent with the greater good depends on the in-
formational environment. It is interesting to consider what the model predicts for
well-established suburban areas with thriving concentrations of retail compared with
low-income neighborhoods without much commercial activity at all. Insofar as sub-
25