This is a guest post by Paul Chaisty and Timothy J. Power. It is based on their paper ‘Flying solo: Explaining single-party cabinets under minority presidentialism’, that is currently available in European Journal of Political Research.
It is now widely acknowledged that presidents whose parties lack majority support in their assemblies attempt to overcome their minority status by building cross-party alliances. Since the onset of the Third Wave of democratisation, presidents in general, and minority presidents in particular have governed with multi-party cabinet coalitions on a frequent basis. Like prime ministers in parliamentary systems, presidents do this through the formation of cabinet coalitions, defined minimally as the awarding of at least one portfolio to a party other than the nominal party of the president. The preponderance of minority presidents and coalition governments has increased as party systems have become more fragmented. Between 1974 and 2013, on average just over half of all minority presidents in political systems that meet minimum democratic standards have governed with multiparty cabinets.
Nonetheless, a large proportion of minority presidents continue to govern with single party cabinets. Whereas 20 per cent of minority prime ministers in parliamentary systems formed their cabinets on a single party basis between 1974 and 2013, unipartisan governments were observed in minority presidential systems almost half of the time (49 per cent). This is puzzling given the many benefits that presidents derive from sharing executive power. Over the last decade, political scientists working almost exclusively on Latin American politics have found for instance that minority presidents who form coalitions increase their legislative productivity (Saiegh 2011) and lower the likelihood of impeachment or removal in times of crisis (Pérez-Liñán 2007).
What explains the adoption of single-party cabinets by minority presidents? In our new article, ‘Flying solo: Explaining single-party cabinets under minority presidentialism’, published last month on-line first by the European Journal of Political Research, we explore this puzzle through cross-sectional time-series analysis of all situations of minority presidentialism in both democracies and semidemocracies between 1974 and 2013. Our analysis covers 610 country-years of minority presidential situations, in which we observe a roughly even split between cabinet coalitions and unipartisan government. Hypotheses are tested that relate to the size and distribution of the formateur (presidential) and largest non-formateur parties that make up the legislature; the nature of party linkages and ideological distance between the president and possible partisan allies; and the extent of reactive veto powers held by the president.
We show that the decision by minority presidents to ‘fly solo’ – that is, to appoint a cabinet made up exclusively of co-partisans – is a function of four main factors: the size of the president’s own party; the concentration of legislative seats in the hands of one non-formateur party; the degree of particularism in the party system; and the institutional capacity of the president to kill or amend unwanted legislation passed by the assembly. Minority presidents who are close to a majority in the assembly, who face a dominant alternative party on the floor, who coexist with party systems in which particularism predominates over programmatic politics, and who possess strong veto powers are significantly more likely to preside over unipartisan governments. Other factors that have been hypothesised to affect presidential strategies – for example the imminence of presidential elections – are found to have little or no effect on this most fundamental of cabinet choices. All of these findings are robust to the inclusion of regional controls.
Of all the factors that we consider in this analysis, the size of the formateur party in the legislature is the strongest stand-alone predictor of single-party cabinets. When all the other variables are held at their means, executives whose parties controlled 49.8 per cent of the seats in the assembly (the maximum value for a minority president under our coding rules) are 47 percentage points more likely to form a unipartisan cabinet than presidents with no legislative co-partisans. We note that this effect takes a linear form: the probability of a non-coalitional outcome increases more or less monotonically in line with the size of the formateur party (see Figure 1).
Figure 1. Predicted probabilities of unipartisan cabinets for four key causal variables, at minimum and maximum values with confidence intervals.
The dominance of a single party over the bloc of non-formateur parties in the assembly is also found to be highly significant. When all other variables were held at their means, those non-formateur parties that controlled 100 per cent of the non-formateur bloc are 46 percentage points more likely to coexist with single-party governments than when the largest non-formateur party held less than 10 per cent of the seats within the non-presidential contingent. Minority presidents operating within party systems characterised by particularistic linkages are also more likely to have single-party cabinets (18 percentage points more likely) than minority presidents facing programmatic party systems. Finally, minority presidents who command strong veto powers are 46 percentage points more likely to form single-party cabinets than presidents with no veto power, holding all other variables at their means.
Therefore, we have moved a bit closer to solving a vexing puzzle about minority presidents. The takeaway message is that presidential authority matters, and it has specific and directional impacts on minority presidents. Their cabinet decisions are affected not only by how close they are to a working majority, but also by the size and salience of prominent nonformateur parties. Presidents whose parties do not control a majority of the assembly are keenly aware of legislative mathematics. These mathematics (i.e. seat distributions) are normally measured by the effective number of parties (Laakso and Taagepera, 1979) or by a Herfindahl fractionalisation index (e.g. Figueiredo et al., 2012).We contend that these aggregate measures are blind to the size and identity of existing political parties, and instead we profile the nonpresidential contingent by measuring the dominance of a single non-formateur party within this bloc. This indicator is far more actor-sensitive than measures of party fragmentation: it captures the relevance of any organised alternative to the party of the incumbent; it can be thought of as a measure of positional rivalry or competition rather than one of dispersion. Simply put, the configuration (as opposed to the fragmentation) of the nonpresidential contingent in the assembly may affect not only the likelihood that invitations to join the cabinet will be issued, but also the probability that these invitations will be accepted.
Our analysis also concurs with recent work that places greater importance on the non-cabinet strategies that presidents use to manage particularistic parties (Kellam 2015; Chaisty and Chernykh 2017). This work shows that presidents may desist from using cabinet powers when forming coalitions in particularistic party systems. Hence, presidents who choose to form single-party cabinets may still form multi-party legislative coalitions in other ways.
Finally, our analysis suggests that the reactive legislative powers of presidents matter. Far from what is implied in a textbook ‘separation of powers’ model, most directly elected presidents around the world have substantial legislative authority, including the power to veto bills either wholly or partially. In this analysis, we find that those minority presidents with strong reactive vetoes are more likely to form unipartisan governments.
Our global, large-N research design trades away some ‘depth’ in return for ‘breadth’. However, the findings here suggest promising avenues of inquiry for presidentialism research in regions where data quality is high and omitted variables can be reinserted (e.g., Latin America), and may help us to establish some parameters for crafting appropriate case study research on the strategic choices of minority presidents.
Chaisty, P. & Chernykh, S. (2017). How do minority presidents manage multiparty coalitions? Identifying and analyzing the payoffs to coalition parties in presidential systems. Political Research Quarterly 70(4): 762–777.
Figueiredo, A.C., Canello, J. & Vieira, M. (2012). Governos minoritários no presidencialismo latinoamericano: Determinantes institucionais e políticos. Dados 55(4):839–875.
Kellam, M. (2015). Parties for hire: How particularistic parties influence presidents’ governing strategies. Party Politics 21(4):515–526.
Laakso, M. & Taagepera, R (1979). Effective number of parties: A measure with application to West Europe. Comparative Political Studies 12(1):3–27.
Pérez-Liñán, A. (2007). Presidential impeachment and the new political instability in Latin America. New York: Cambridge University Press.
Saiegh, S.M. (2011). Ruling by statute: How uncertainty and vote buying shape lawmaking. New York: Cambridge University Press.