The responsive legislature: Public opinion and law making in a highly disciplined legislature

British Journal of Political Science

Volumn 37, Issue 2, 2007, Pages 263-280

  Calvo, Ernesto ^a  

^a UNIVERSITY OF HOUSTON   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33947699908     PISSN: 00071234     EISSN: 14692112     Source Type: Journal    
DOI: 10.1017/S0007123407000130     Document Type: Article

References (52)

1. See Jon R. Bond, Richard Fleisher and B. Dan Wood, 'The Marginal and Time-Varying Effect of Public Approval on Presidential Success in Congress', Journal of Politics, 65 (2003), 92-110;
2. Brandice Canes-Wrone, 'The Public Presidency, Personal Approval Ratings, and Policy Making', Presidential Studies Quarterly, 34 (2004), 477-92;
3. Brandice Canes-Wrone and Scott de Marchi, 'Presidential Approval and Legislative Success', Journal of Politics, 64 (2002), 491-509;
4. Samuel Kernell, Going Public: New Strategies of Presidential Leadership (Washington, D.C.: CQ Press, 1986). There is considerable disagreement, however, with respect to the magnitude of this effect.
5. John Carey and Matthew Shugart, Presidents and Assemblies: Constitutional Design and Electoral Dynamics (New York: Cambridge University Press, 1992);
6. Scott Morgenstern, 'Explaining Legislative Politics in Latin America', in Scott Morgenstern and Benito Nacif, eds, Legislative Politics in Latin America (New York: Cambridge University Press, 2002), pp. 413-46;
7. Scott Morgenstern, Patterns of Legislative Politics: Roll Call Voting in Latin America and the United States (New York: Cambridge University Press, 2004);
8. Gabriel Negretto, 'Government Capacities and Policy Making by Decree in Latin America: The Cases of Brazil and Argentina', Comparative Political Studies, 37 (2004), 531-62.
9. Mark P. Jones and Wonjae Hwang, 'Party Government in Presidential Democracies: Extending Cartel Theory Beyond the US Congress', American Journal of Political Science, 49 (2005), 267-82.
10. Gary W. Cox and Mathew D. McCubbins, Setting the Agenda: Responsible Party Government in the US House of Representatives (New York: Cambridge University Press, 2005). Notice that if the representatives are perfectly disciplined on the floor, ideal point estimates obtained from roll-call votes will not discriminate among different representatives of the cartel. Therefore, although the preferences of the representatives are crucial for explaining legislative success, these preferences are not directly observable to the researcher. To overcome this problem, Jones and Hwang code absentee votes as 'nay' votes, therefore adding some variance to the estimated preferences of representatives in the Argentine Congress. This, however, does not solve the most significant problem, which is that preferences about legislation that has been prevented from reaching the floor are never measured. Failure to get legislation approved, however, provides indirect information about the preferences of legislators.
11. Bills can be initiated by a senator, a deputy or the executive. The executive's decision to initiate legislation in the House or the Senate is based on strategic calculations resulting from the procedural advantages enjoyed by the initiating chamber. Records of all proposed legislation can be obtained from Información Parlamentaria (www.diputados.gov.ar). Most studies of presidential legislative success in the United States, by contrast, measure the publicly recorded position of the president with respect to particular bills introduced by legislators. This may lead to potential biases, given that public support for a particular bill does not immediately indicate that the president is committed to the proposed legislation (strategic position taking).
12. Cox and McCubbins, Setting the Agenda, p. 41.
13. See Jones and Hwang, 'Party Government in Presidential Democracies'. The authors provide a similar argument, with governors taking the role of senior partners in charge of distinctive provincial delegations. Ideal point estimates obtained from floor-votes, however, fail to find support for voting differences across provincial delegations. This article shows that evidence of the preferences of senior partners should be observed in the agenda, rather than the vote. At the aggregate level, evidence of disagreement among senior partners should be visible in the variations in bill approval rates rather than in the nominal votes for particular pieces of legislation.
14. Cox and McCubbins, Setting the Agenda, p. 226.
15. Cox and McCubbins, Setting the Agenda, p. 114.
16. Adam Przeworski, Susan C. Stokes and Bernard Manin, Democracy, Accountability, and Representation (New York: Cambridge University Press, 1999).
17. A nice example can be found in the implementation of the value-added tax reform voted by the Argentine Congress. See Kent Eaton, 'Fiscal Policy Making in the Argentine Legislature', in Morgenstern and Nacif, eds, Legislative Politics in Latin America, pp. 287-314.
18. Mark P. Jones, Sebastian Saiegh, Pablo T. Spiller and Mariano Tommasi, 'Amateur Legislators-Professional Politicians: The Consequences of Party-Centered Electoral Rules in a Federal System', American Journal of Political Science, 46 (2002), 356-69.
19. David Mayhew, Divided We Govern: Party Control, Lawmaking, and Investigations, 1946-1990 (New Haven, Conn.: Yale University Press, 1991).
20. William Howell, Scott Adler, Charles Cameron and Charles Riemann, 'Divided Government and Legislative Productivity of Congress, 1945-1994', Legislative Studies Quarterly, 25 (2000), 285-312;
21. Keith Krehbiel, Pivotal Politics: A Theory of US Lawmaking (Chicago: Chicago University Press, 1998).
22. Cox and McCubbins, Setting the Agenda, p. 114.
23. Eaton, 'Fiscal Policy Making in the Argentine Legislature'.
24. Side payments could also be critical, as became clear with the bribery scandal in the Senate during De La Rua's administration.
25. See Mariana Llanos, Privatization and Democracy in Argentina: An Analysis of President-Congress Relations (New York: Palgrave Macmillan, 2002). The data collected for this project also shows that at 58 per cent of presidential initiatives were also approved by at least one of the chambers.
26. The rate of approval of presidential initiatives when we control for procedural advantages at the floor level. In the model we describe in the third section, the effective approval rate is captured by the random intercept of the hierarchical model. Figure 1 plots the intercept of a dynamic hierarchical model using a time ticker over groups of twenty bills. See Appendix A. The bills under study include all significant legislation entered in Congress, and excludes formal bills such as requests to leave the country (official travel), bills to ratify co-operation and good will treaties, and confirmation of cabinet members in their posts.
27. The multipartidaria was a committee created in Argentina in 1981 by the five most important political parties to negotiate the democratic transition process. The formation of the multipartidaria was instrumental in bringing light on a secret pact between the Peronist presidential candidate Italo A. Luder and the Military Junta. This secret pact was one of the most salient issues debated by the presidential candidates in the 1983 campaign, leading to the election of the UCR candidate Raúl R. Alfonsín.
28. In the Argentine case, however, Peronists have controlled a majority of the Senate since 1983. Therefore, it was enough for the Peronists to win the House to control the amendment and approval process entirely. The rate of approval of presidential initiatives could depend on the number of seats held by the incumbent president but it could also be affected by the number of lower chamber seats controlled by the Peronists, given that larger Peronist House minorities could lead to the approval of heavily amended presidential bills even under non-Peronist presidents. As a result, an increase in the number of executive seats could lead to an increase in the bills approved by Congress in a model of executive dominance, while, by contrast, an increase in the number of lower chamber Peronist seats increased the approval rate of amended presidential bills under non-Peronist executives. In the final section, I test both models and show that both are weak predictors of bill approval.
29. Jones and Hwang, 'Party Government in Presidential Democracies'.
30. Most failed initiatives were never sent to committee and were never brought to the floor. Executive proposals that were rejected (Rechazado), withdrawn (Retirado), or filed without being considered (Archivado), add up to less than 1.5 per cent of all projects.
31. Eduardo Aleman and George Tsebelis, 'The Origins of Presidential Conditional Agenda Setting Power in Latin America', Latin American Research Review, 40 (2005), 3-26;
32. Eduardo Aleman and George Tsebelis, 'Presidential Conditional Agenda Setting Power in Latin America', World Politics, 57 (2005), 396-420.
33. The code and data to run this model may be requested from the author or downloaded from http://calvo.polsci.uh.edu/code.htm.
34. Notice that the random intercept has been logistically transformed in the first level. Therefore, the second-level estimate of a should be interpreted as the log-odds mean bill approval ratio and the second-level equation follows a normal distribution.
35. For a general introduction to Bayesian statistical models, see Andrew Gelman, John B. Carlin, Hal S. Stern and Donald B. Rubin. Bayesian Data Analysis (Boca Raton: Chapman & Hall, 2000).
36. For example, to model confidence intervals around the mean presidential approval rating in the second-level equation.
37. Andrew Gelman and lain Pardoe, 'Bayesian Measures of Explained Variance and Pooling in Multilevel (Hierarchical) Models' (New York: Columbia University working paper, 2004).
38. David Spiegelhalter, Andrew Thomas, Nicky Best and Dave Lunn, WinBUGS User Manual 1.4 (Cambridge University, 2003), available at http://www.mrc-bsu.cam.ac.uk/bugs. All replication material can be found in http://calvo.polsci.uh.edu/code.htm.
39. Unlike in the United States, the confirmation of presidential appointees in the Argentinian Congress is mostly a formal procedure.
40. The date the bill was initiated is used to indicate the congressional year (or period).
41. Legislation can be initiated during ordinary congressional sessions or during extended sessions requested by the executive (sesiones extraordinarias).
42. While the honeymoon variable should be considered an aggregate-level variable, it does not always coincide with the calendar year. For example, Carlos Menem took office in mid-July of 1989. For all other aggregate-level variables, I divided the year of 1989 into two separate years, but because the honeymoon variable includes legislation introduced in the first semester of 1990, it does not match all other aggregate-level variables. Adding a separate aggregate level in the model complicated the estimation unnecessarily. Therefore, honeymoon was included as a first-level variable.
43. If the bill was initiated in the House and never made it to the Senate, the variable indicating the number of Senate committees was set to that year's mean. Similarly, if the bill was initiated in the Senate and never made it to the House, the value of the number of House committees was set to that year's mean. This adjustment is necessary to prevent the absence of information in the House or the Senate to bias the estimates of the model. I thank one of the anonymous reviewers for noticing a problem in the coding of this variable.
44. As noted by an anonymous reviewer, the model specification is taking the effect of the first-level independent variables as fixed across years. An alternative specification could allow these estimates to vary by Congress, under the presumption that the slope of the relationship is also affected by contextual variables, as described by Bond, Fleisher and Wood, 'The Marginal and Time-Varying Effect of Public Approval on Presidential Success in Congress'. In that article, the authors find that the effect of public approval on legislative success is sporadic or episodic, and propose an alternative frequentist specification interacting public approval with partisanship (a variable measuring the degree of party polarization on the vote). Running an equivalent Bayesian specification for the Argentine Congress, however, did not detect time-varying effects. Two reasons that may influence such a non-finding are the shorter time-series run for the Argentine dataset and the high level of discipline in floor votes as described earlier.
45. Other independent variables tested but not introduced in the final model include change in gross domestic product and a dummy indicating divided government. They were not introduced for theoretical reasons. When introduced in the model the results were not substantively or statistically significant. See Appendix A for a description of the variables and descriptive statistics. We also replicated these analyses using a variable measuring the positive image of the president's economic policy. The results can be requested from the author.
46. As indicated in fn. 34, the Congress of 1989 was divided into two, due to the early resignation of Raúl R. Alfonsín. Therefore, there are nineteen Congresses but twenty aggregate-level observations.
47. The periodic public opinion data is irregularly spaced over time. This does not constitute a problem for the analyses given that it is an independent variable. The expected first level variance of second and third models is higher, due to the smaller number of bills subject to treatment in each of the eighty-six periods.
48. The fully Bayesian specification with eighty-six second-level observations can be requested from the author.
49. The results are robust to many different specifications including general linear models with logit and probit links, with and without random effects. As expected, the results presented in Table 1 provide the most conservative estimates.
50. The classification of different types of legislation follows the design of Michelle Taylor-Robinson and Christopher Diaz, 'Who Gets Legislation Passed in a Marginal Legislature and is the Label Marginal Legislature Still Appropriate? A Study of the Honduran Congress', Comparative Political Studies, 32 (1999), 589-625.
51. With all values set to their means and presidential approval set to a low of 10 per cent or a high of 90 per cent.
52. Robert S. Erikson, M. Mackuen and J. Stimson, The Macro Polity (New York: Cambridge University Press, 2002).