1.2. DEFINITIONS
35
flexible and responsive, but not overreact to economic, political, and social challenges. This
contrasts with a federation in stasis, where the analysis does not take into account the
federation’s capacity to evolve or respond.
A federation’s performance is only as good as the inputs: the recipe—the distribution of
authority—and the federal and state governments’ compliance with that distribution. If the
distribution is flawed, then the federation will not perform well, no matter how compliant the
governments are. It can be flawed for a number of reasons. First, we must acknowledge our
fallibility as social scientists, and the limits of our understanding as designers. As Chapter
2 will describe, designing the allocation of authority is a great problem in social engineer-
ing. People are not atoms; their actions and reactions surprise the institutional engineer. A
perfect design would demand a perfect understanding how people will react to complex, in-
terdependent incentives, but we have only an imperfect understanding about the relationship
between the distribution of authority and the union’s ability to reach its potential. Second
is the trade off in any distribution. In section 2.6 I make the case that not all objectives are
complementary; pursuit of some compromise a union’s ability to pursue others. If the union
is evaluated only along the dimension that is sacrificed, it will appear lackluster. Third is
a natural extension of the second: with heterogeneity in the population, some will prefer
one distribution over another because of the asymmetric consequences. Subgroups within
the population would rank order potential distributions of authority differently, depending
upon the subgroup’s preferences. Fourth further extends this asymmetry: the adoption of
the distribution of authority may be affected by power asymmetries.
The second through fourth points underscore the delicacy of adaptation. To a large
extent, the political economy of institutions assumes that players willingly enter into the
environment established by institutions—that play is voluntary. In an overview of the power
approach to analyzing institutions, Moe (2005) warns against an overly rosy view of insti-
tutions, since players subject to an institution’s incentives may be forced to play according
Convention