34
Usually overlooked by the literature18, declarations, resolutions and
conununications have not been included in the samples of most analyses, ɪn fact, Jones &
Hwang's analysis is based on estimations of roll call votes. No one doubts the relevance
of roll call voting, but the sample of bills that actually goes to roll call is extremely low19.
The mentioned piece has a sample of 473 bills between 1989 and 2003, while about
26,000 national bills have been treated and about 2,000 have been passed dining that
period. If declarations, resolutions and communications are also counted, the number of
bills introduced reaches 78,000, and about 30,000 of these are approved. In Calvo's
study, only the set of presidential bills are used to test his hypotheses. Even though the
sample is fully adequate for his research goals (rate of presidential bills' approval), it
only considers 2,969 of the 177,302 bills that have been submitted to the Argentine
Congress between 1983 and 2007. Far from thinking that the choice of the samples
invalids the findings of the aforementioned pieces; I do believe that a comprehensive
depiction of the workings of the Argentine Congress requires an analysis that includes
the full sample of legislation introduced in the period.
The use of all the kinds of legislation can be useful for several goals. First,
looking at the whole picture can be better forged; in as much, it avoids any kind of
selection bias in the choice of the sample. Second, it lets me reconcile the party-level
arguments of the previous literature with an individual-level approach that has been
overlooked until now. Analyzing all kinds of legislation will contribute to the
understanding of congressional performance in a case where subjects are strongly
between governors (and sometimes mayors) and officers of the executive branch (ministers or secretaries of
Infrastructure, Finances or Internal Affairs). See Tommasi (2002) and Spiller and Tommasi (2007) for a
good discussion.
8 Calvo (2007) calls them “minor bills with a symbolic value”
*9 See Carubba et al (2006) for a good discussion about use of roll-call and selection biases.