Executive-Legislative Deadlocks in the Dominican Republic

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

If we look at each presidency, we see that Antonio Guzmán's presidency (PRD, 1978-82) had relatively high activity compared to that of Salvador Jörge Blanco (PRD, 1982-86), with a mean production of important laws per legislature of 11.6 and 8.8, respectively. These presidencies have been described as deadlocked (e.g., Hartlyn 1998; Dario Espinal 2001; Jiménez Polanco 1999), Guzmán's presidency because the PRSC had control of the Senate, and Jorge Blanco's (majority) presidency because of a formal split of the incumbent PRD. Nevertheless, when compared to the Balaguer presidencies of 1986-96, generally regarded as strong, with a mean production of laws per legislature of 6.5, both preceding PRD presidencies are more active.

This also holds when we compare the production of all laws. Even though the PRD presidencies of 1978-86 experienced tremendous problems implementing their political agenda, these data indicate that the Guzmân administration was not as deadlocked as previously expected. Guzmán got all his budgets passed in Congress, and, as the only president in the 1978-2005 period, he even got two of his budgets passed during the ordinary fall legislature. All other presidents since 1982 have had to convene either a prolonged or an extraordinary legislature to get their budget passed; and three times in the period the budget failed to pass (Jorge Blanco in 1985, Balaguer in 1990, and Fernández in 1996). This indicates that President Guzmán was able to get out the votes in Congress. Furthermore, when compared to the reform-eager presidency of Fernández (PLD, 1996-2000), with a mean production of 13-3 important laws per legislature, the PRD presidencies do not stand out for the low production of important laws.

Executive-Legislative Relations as a Continuous Variable

To analyze whether our hypotheses can predict and explain the occurrence of deadlocks, we look first at the executive-legislative relations as a continuous variable, measured by the number of important laws per legislature. Table 1 shows the means and the standard deviations of laws produced per legislature in the Dominican Republic in the 1978-2005 period, according to the four hypotheses discussed above and broken into periods in which deadlocks would be expected and not expected. Except for the effective number of parties, which are ordered into four values, the independent variables are dichotomous. The table also presents the F-values of a one-way ANOVA difference of means test and their respective levels of significance. The period measured consists of 52 legislatures, excluding the same three legislatures as reported above. The overall means of important laws per legislature and standard deviation are 9.8 and 6.5, respectively.

All periods when deadlocks are predicted produce, as expected, on average fewer important laws per legislature than when the various hypotheses do not predict a deadlock, and fewer than the overall mean for the whole period (compared to the value 9.8). However, none of the F-values for the four hypotheses reaches the normal significance level of .05.16 Only Cheibub's veto hypothesis is close, at .053. Thus our institutional variables do not affect the production of important laws in the Dominican Republic in a significant way.