Executive-Legislative Deadlocks in the Dominican Republic

Latin American Politics and Society, Summer 2008 by Marsteintredet, Leiv

The question of what should be the appropriate time unit for studying the occurrence of deadlocks is a dilemma. This dilemma relates to all attempts to measure executive-legislative relations and deadlocks but, nevertheless, is not discussed in the literature. A normal approach in the literature is to use a president's administration as the appropriate time unit. This is problematic because a deadlock might last only a year or even less, and applying a measure of deadlocks with a four-year administrative period as the time frame could thereby underscore the occurrence of deadlocks. Furthermore, executive-legislative relations are never constant for a full electoral period. Therefore, this study uses a legislature as the appropriate time unit for measuring executive-legislative relations.12

Lacking data on vetoes or data that enable us to use the box score, we suggest a count of important or relevant laws as a very crude proxy to measure executive-legislative relations. Whether the president vetoes legislation or Congress refuses to pass it, we would expect a drop in the number of laws passed during a legislative period, regardless of the laws' importance. The problem is, of course, how to define an important law, even though this problem is more acute when comparing countries than in a single case study. The count of important laws in this case is the sum of all laws that are not defined as nonrelevant, plus all highly important laws, such as national budgets, constitutional reforms, laws that regulate the relationship between state institutions, and laws that relate to the three democratic features: participation, contestation, and civil liberties and human rights. The nonrelevant laws that are excluded are private pensions and symbolic and insignificant laws, such as the naming of streets, the issuing or memorial coins, and the like.13 This excludes 1,230 of 1,758 laws produced in the period 1978-2005. Of the remaining 528 laws that constitute our data material, 44 were coded as highly important.

The suggested measure encounters several validity problems. The most serious problem is that the count of laws neglects the input side of legislation; that is, the number of bills initiated either by the president or Congress, often called the denominator problem. The continuous variable is a measure of productivity rather than deadlocks, and does not capture whether the drop in legislation is due to few initiated bills, natural variation of productivity across administrations, or a deadlock.14 The qualitative analysis and the dichotomous operationalization of deadlocks attempt to control for these problems.

Executive-Legislative Relations as a Dichotomous Variable

Executive-legislative relations until now have been generally understood as a graded concept. In our view, the phenomenon of deadlocks can equally well be captured as a dichotomous concept: either there is a deadlock or there is not. In line with Collier and Adcock's (1999) pragmatic approach to concepts, this study argues that given the data available and the goals of the research, a dichotomous understanding of executive-legislative relations and deadlocks is the better choice. A dichotomous concept can control for some problems with the above operationalization, but two problems arise: what is the appropriate baseline for calculating the cutoff point? and what is the appropriate deviation from the baseline for calculating the cutoff point for deadlocks?