II. The model
2.1. Basic assumptions
Let us consider a population with the size normalized to unity that consists of two
groups: the bosses and the workers. The bosses’ incomes are essentially political
(bureaucratic) rents. Individual rents are identical and equal R. The bosses retire after Tb years
in office. The rest of the population, the workers and the retired bosses, earn a uniform wage,
W, which is substantially lower than the political rent: W <<R. A part of the working
population, the activists, provide additional services that benefit the bosses by increasing their
rents: R = f(Na) > R0 > 0, where f(Na) is the activists’ “rent-production function” such that
f (Na) ≥ 0 and f'(Na) ≤ 0; and R0 is the bureaucrats’ potential per capita rent in the absence of
activists.9 R0 and W are positive and assumed exogenous. The bosses and the activists
comprise small fractions of the total population: Nb << 1,10 Na << 1. No specific assumptions
are made with respect to the ratio of the numbers of bosses and activists.
Activist service requires extra effort on the part of the workers who choose to enlist as
activists. No immediate compensation is offered for the service. Each activist volunteers a
unit of extra effort. However, the workers are heterogeneous in their idiosyncratic distaste for
activist service - the disutility of the extra effort that the activist service calls for. This
disutility has an income equivalent of ω i, which is an inverse measure of an individual’s
predisposition (or ability) for activist service. It is distributed within the population with c.d.f.
9 R0 measures the bureaucrats’ own productivity in rent-collection. R0 is generally greater than zero,
since a rational worker will not fight corruption or resist predatory taxation beyond the level that
equates the losses form the latter and the costs of enforcing the rule of law at the margin.
10 For all practical purposes, the bosses can be assumed not to exceed one percent of the population.
12