Legislative Fragmentation and Government Spending in Presidential Democracies: Bringing Ideological Polarization into the Picture

• Author: Oskar Nupia,Marcela Eslava
• Date: 01 August 2017
• Published date: 01 August 2017
• DOI: http://doi.org/10.1111/lsq.12152

MARCELA ESLAVA

OSKAR NUPIA

Universidad de los Andes

Legislative Fragmentation and

Government Spending in

Presidential Democracies:

Bringing Ideological Polarization

into the Picture

We claim that, in presidential democracies, the effect of increasing fragmentation

on government spending should be conditional on polarization, defined as the ideological

distance between the government’s party and other parties in Congress. We buil da mod-

el where this result follows from negotiations between the legislature and an independent

government seeking the approval of its initiatives—as in presidential democracies. Using

cross-country data over time, we test the empirical validity of our claim finding that, in

presidential democracies, there is indeed a positive effect of fragmentation only when

polarization is sufficiently high. The same is not true for parliamentary democracies.

I need to engage in give-and-take, build consensus,and engage in negotiations without

sacrificing principles.

(Indonesian President Susilo Bambang Yudhoyono,

interview airedon CNN, June 15, 2011)

Considerable attention has been given in the literature to the idea

that fragmentation affects f‌iscal discipline (e.g., Bawn and Rosenbluth

2006; De Haan, Sturm, and Beekhuis 1999; Edin and Ohlson 1991;

Elgie and McMenamin 2008; Perotti and Kontopoulos 2002; Pettersson-

Lidbom 2012; Schaltegger and Feld 2009; Volkerink and de Haan 2001;

Weingast, Shepsle, and Johnsen 1981). Fragmentation in this context

refers generically to the number of different interests whose demands are

ref‌lected in the government’s budget. Greater fragmentation is expected

to relax f‌iscal discipline through a standard “common pool of resources”

problem. The greater the number of groups benef‌iting from government

expenditure and sharing its costs, the lesser the degree to which any

LEGISLATIVE STUDIES QUARTERLY, 42, 3, August 2017 387

DOI: 10.1111/lsq.12152

V

C2016 Washington University in St. Louis

particular group internalizes the costs of the goods its members receive

from the government. As a result, the total amount of spending

demanded by participating groups grows with their number. This idea

was put forward a long time ago (e.g., Weingast, Shepsle, and Johnsen

1981) and has since been the subject of a large literature.

A good part of that literature has focused on testing empirically

whether a specif‌ic source of fragmentation, “legislative

fragmentation”—the number of legislating parties in a governing coali-

tion—leads to larger public spending or greater def‌icits.

1

This is also the

focus of our article. The f‌irst wave of studies on this issue concentrated

on samples of countries that have been part of the OECD since its incep-

tion, f‌inding a positive effect of fragmentation on government spending

(e.g., Bawn and Rosenbluth 2006; Perotti and Kontopoulos 2002;

Volkerink and de Haan 2001) or f‌iscal def‌icits (Perotti and Kontopoulos

2002; Volkerink and De Haan 2001). While the results mentioned above

are consistent with the basic thesis that greater fragmentation implies

less f‌iscal discipline, a second wave of studies covering both OECD and

non-OECD countries has found a weaker effect of legislative fragmenta-

tion on f‌iscal balances than that found for the OECD alone (Elgie and

McMenamin 2008; Mukherjee 2003; Woo 2003).

We suggest here that the reason why the link between fragmenta-

tion and f‌iscal discipline appears weaker in non-OECD samples of

countries than in OECD ones may have to do with the predominant pres-

ence of presidential democracies in the former samples and of

parliamentary ones in the latter. As Hankla argued recently, “although

this is rarely acknowledged in the literature, the theoretical link between

party fragmentation and def‌icit spending implicitly assumes a parliamen-

tary system. In presidential systems, the executive is not drawn from the

legislature, and so fragmented assemblies do not translate automatically

into fragmented governments. As a result, there are very good reasons to

believe that the linear relationship between divided legislatures and bud-

get def‌icits may not hold for presidential democracies” (2013, 200). In

particular, the process of ongoing negotiation that characterizes the rela-

tionship between the government and the legislature in a presidential

democracy is absent from the aforementioned theoretical argument. Fac-

tors inf‌luencing that negotiation process may in the end intermediate the

relationship between fragmentation and the budget in presidential

democracies. In this article, we highlight one such factor: the degree of

ideological dispersion between parties that the government negotiates

with, termed here “polarization.”

Once this dimension is considered, the fragmentation-spending

link in presidential democracies should be signif‌icant only in the

388 Marcela Eslava and Oskar Nupia

presence of high enough polarization. The reason is straightforward: In a

context where pork barrel is used by an independent government to

obtain support for its public policies in the legislature (as is the case pre-

dominantly in presidential systems), low polarization around these

policy initiatives implies that the government can obtain legislative sup-

port paying little attention to the coalition parties’ demands for pork. The

common-pool problem imposed by these demands is thus minimized.

However, if there is high polarization around these policy initiatives, the

government party must satisfy at least some of these demands in order to

get legislative approval. In this case, the degree of political fragmentation

should positively affect the level of public spending.

To test our hypothesis that fragmentation matters only in the

presence of high enough polarization, we use information for a panel of

presidential democracies for the period 1978–2005. We measure frag-

mentation based on the number of legislating parties in the governing

coalition and polarization based on the ideological distance between

them (the details are discussed below). We focus on the effects fragmen-

tation and polarization have on central government spending, as

opposed to def‌icits. As noted by Perotti and Kontopoulos (2002), the

theoretical arguments outlined above link fragmentation with govern-

ment spending; arguments translating the pressure derived from

fragmentation into greater def‌icits are less general. It is worth highlight-

ing that fragmentation and polarization, though likely correlated, are

different concepts, as def‌ined above.

Our f‌indings support our hypothesis. We f‌ind that the effect of

fragmentation on government spending in presidential regimes is

increasing in polarization and is signif‌icantly different from zero

(either in a statistical or economic sense) only when there is a high

enough degree of polarization. By contrast, when extending the exer-

cise to a panel of parliamentary democracies, we f‌ind that

fragmentation has an unconditional (i.e., not intermediated by polari-

zation) positive and signif‌icant effect on spending for these countries.

Moreover, our argument and evidence are consistent with previous

f‌indings showing that the direct effect of fragmentation on govern-

ment spending holds for parliamentary democracies but not so in

samples of presidential ones (Mukherjee 2003).

Other recent analyses have studied how institutional and political

differences across types of democracy affect the relationship between

fragmentation and f‌iscal policy (Cheibub 2006; Hallerberg and Marier

2004; Hankla 2013; Wehner 2010). Cheibub (2006) shows that budget

discipline varies between presidential and parliamentary democracies

and that within presidential democracies stronger presidents keep more

389Bringing Ideological Polarization into the Picture