makers then lack the willingness to reform.
Moreover, individuals may have preferences about the way institutions actually should be
designed. It is obvious that these preferences are shaped by either the former institutional
system or severe economic shocks and that historically shaped preferences do have an effect on
current political outcomes.
De Grauwe (2004), for example, argues that historical events explain the differences between
the French and German inflation target prior to the introduction of the European Monetary
System (EMS). Severe inflationary crises in the German history and painful deflationary episodes
in the French have shaped preferences among politicians and citizens in favour of more stability
oriented or lax monetary policy respectively.
Similarly, Alesina and Fuchs-Schiindeln (2007) find that East Germans, who grew up under
the socialistic regime, are more in favour of redistribution and state intervention than West
Germans. This effect is especially strong for older cohorts, which have spent a longer time of
their lives under the Communist regime.
The education system is also likely to be another factor causing persistence of institutions
in Central Asia. Sound property rights protection and well functioning factor markets have
been found essential preconditions for developing a knowledge-based economy and economic
growth driven by ideas. However, having a well developed property rights protection alone does
not ensure producing ideas, as it requires a certain level of education among the work force to
develop innovations.
Figure 3 shows third level education enrollment rates for Central Asia and the CEEC. Look-
ing at Figure 3 shows that there seems to be no connection between education and institutions
around 1995. Kazakhstan and Uzbekistan lead the figures followed by the CEEC, Turkmenistan,
Tajikistan, and the Kyrgyz Republic in the last spot. However, third level education enrollment
rates roughly mirror the ranking of institutional quality in Figure 1 when looking at 2006 fig-
ures. CEEC have the highest share of population with third level education in 2006 followed
by Kazakhstan and the Kyrgyz Republic as the countries with the highest share of third level
education among Central Asian countries, and Turkmenistan and Uzbekistan with the lowest
fraction.
Hence, while education did not seem to matter much at the beginning and mid-1990s ed-
ucation and knowledge of the labour force seem to be associated with the reform process in