Call Us: US - +1 845 478 5244 | UK - +44 20 7193 7850 | AUS - +61 2 8005 4826

Macroeconomic analysis

It is easy enough to define a threshold y* that is sufficient to generate all the effects above. Note that the highest probability of an attack is
bounded above by p, the probability of a cross-religious confrontation.
If, at this level, it is optimal for an individual to choose the “human protection” technology, then by the first part of observation 1, it is optimal
to do so for all lower levels. It is straightforward to see that such a threshold must exist.12
For individuals with incomes that exceed this threshold, the capitalintensive technology may be attractive. If it is attractive both before and
after the change in group incomes, then the effect on d will depend on
the ratio of w* to y. If w* is a fully human cost and involves the use of
fellow group members, it will again be proportional to group means,
and previous arguments apply. The ambiguity arises from individuals
whose incomes cross the threshold. Figure 3 shows what happens with
incomes that rely on the human technology before the change but move
into the fixed-cost technology after the change. Panel A shows that it is
now possible for there to be a sharp upward jump in defense expenditures.13 The protection function shifts downward, as in panel B, while the
attack function ðas beforeÞ shifts upward. The net effect will depend on
the relative strengths of these two shifts, and it is ambiguous.
The effect on overall attacks will depend on the proportion of individuals who fall below the threshold for which the capital-intensive
technology is never used. The more individuals there are in this category,
the more likely it is that economic improvement