opportunities for vertical income mobility which was not introduced explicitly in the
theoretical model. The negative effect of this variable is in agreement with this interpretation.
It can be argued, however, that the ruling bureaucracy values an educated ‘cadre’ and,
therefore, higher education increases the chances for promotion and/or is one possible sort of
reward for the activists. This contradicts the alternative-opportunity function of higher
education. The low significance of ST in most specifications can be, therefore, considered as
resulting from the counteraction of the two effects of the higher education on the incentives to
join the party. Finally, significant negative estimates of the After_1961 dummy can be
explained by the elevated expectations of promotion on the part of the activists due to the
temporary increase in the rate of turnover within the bureaucracy in 1956-61, when Stalin’s
cohort of bosses was largely forced to retire.
A distinctive feature of the results is the higher significance of variables that
correspond to the negative stimuli to join the party (retail sales, enrollment in higher
education, etc.) vs. low significance of positive ones (the number of bosses and their average
salary). The contribution of the latter into the explained variation is one order of magnitude
lower than that of the former. Although the proxy for bureaucratic rents used here - salary -
accounts for only a portion of the total remuneration of the bureaucrats, there might be a more
general explanation for the relative strength of the “push” to seek for political careers, as
opposed to the “pull” of expected rents. The variables that determine the latter are far from
perfectly observable for a worker considering the choice to become an activist. The
information on the number of positions in the bureaucracy and the bosses’ salaries and
benefits is hardly public domain under any non-democratic regime. It could be acquired
37