2.3 Unresolved questions
The existing studies leave a number of questions open. Most of the literature merely looks at
fiscal policy in a highly aggregated way, restricting empirical tests largely to the budgetary
balance and ignoring specific developments on both sides of the budget. Furthermore, a precise
analysis of the time profiles of reform related fiscal effects is missing.
A weakness of dominating descriptive approaches is that they do not distinguish between changes
within the course of reform cycles on the one hand and differences between stable periods and
reform cycles on the other hand. In terms of the stylized setting above a comparison between
typical characteristics of R periods with all other periods is not very informative. Instead it is
desirable to understand, firstly, which features are typical for the transition from S to PR and,
secondly, how economic and fiscal variables evolve from PR to SA. Straightforward comparisons
between reform and no-reform periods mix these very different issues up and, hence may lead to
misunderstandings.
A further shortcoming is the unclear explanation for the frequent finding of increasing deficits
with certain types of reforms. Although the compensation argument is regularly cited in view of
this correlation none of these studies is really able to make an empirical distinction between the
different possible explanations which, besides compensation, could also be related to direct fiscal
implication of reforms or to expectations effects.
Generally, expectation and immediate demand effects are a completely neglected aspect of this
empirical literature so far and require further attention.
3 Time profiles and properties of reform periods
3.1 Data and approach
Reforms are measured as changes in the restrictiveness of government regulation of product,
labour and financial markets and a measure for the degree of distortions associated with the
country’s tax system. Data on regulatory rigidity originate from the data collection used in the
139