The eleven propositions laid out in the previous chapter necessitate a number of data sets for empirical testing. In this chapter, I explain how I have produced a data set of major EU laws and derived measures of discretion for the Commission and national authorities. These will be the core dependent variables for testing the first eight hypotheses.

In the first two sections, this chapter outlines the procedure employed for the selection of major laws and the sample of laws that has been used for the empirical analysis. It then proceeds to explain the steps that I took to produce the measures of discretion. I make ample use of descriptive statistics and longitudinal analysis to provide broad impressions of the data set and to assess the validity of existing claims on the historical development of the EU.

The selection of major EU laws

The procedure for selecting major EU laws should be systematic and minimize bias. It should also employ a neutral method to separate significant from trivial legislation. In compiling a list of important post-war US legislation, Mayhew (1991) wanted to capture both retrospective and contemporary judgements on law production. He used references in scholarly books and year-end roundups of both the New York Times and Washington Post. To my knowledge, there are no lists available of major EU laws, yearly commentaries or publications that are selective enough and can be used systematically from 1958 onwards.