Content uploaded by Ignacio Arana
Author content
All content in this area was uploaded by Ignacio Arana on Nov 21, 2017
Content may be subject to copyright.
Informal Institutions and
Horizontal Accountability:
Protocols in the Chilean Budgetary Process
Ignacio Arana Araya
ABSTRACT
Studies of executive-legislative relations are usually based only on the analysis of
formal institutions, although informal institutions also shape interbranch behavior.
This omission leads to questionable results when scholars examine the capacity of
state institutions to audit other public agencies and branches of government. This
article explores how the protocols, an informal institution that shapes the Chilean
budgetary negotiations, have increasingly allowed the congress to have a more rel-
evant budgetary role than what the constitution permits. It argues that protocols
accommodate some of the undesired consequences of a charter that is strongly
biased toward the central government, and describes how this institution has
departed from its stringent budgetary focus to encompass broader executive-legisla-
tive agreements that enhance the legislature’s capacity to oversee the executive.
In an influential article, Guillermo O’Donnell (1998) argues that many develop-
ing democracies lack horizontal accountability: allegedly autonomous state
powers are unable to exercise controls over other public agencies (see also Mainwar-
ing and Welna 2003). Weak checks among state powers lead to violation of laws,
less transparency, more corruption. and an underrepresentation of the preferences
of the voters that elect the congress. O’Donnell claims that the lack of horizontal
accountability could be explained by the patterns of interaction between formal and
informal institutions. However, although informal rules shape politicians’ behavior
(see, e.g., Helmke and Levitsky 2006), studies of executive-legislative relations usu-
ally consider only formal rules.
This study examines the role of protocols (protocolos), agreements signed
between the two elected branches of government during the budgetary negotiations
in Chile. It argues that the protocols are an informal institution that bypasses the
1980 Constitution and helps hold the government accountable to Congress. Proto-
cols commit the executive to disclosing information, making concessions, commis-
sioning research, evaluating programs, sending bills to Congress, and increasing,
limiting, and reallocating expenditures.
Previous studies have examined informal politics in the Chilean executive-leg-
islative relationship (e.g., Ferraro 2008; Siavelis 2006; Carey and Siavelis 2006;
© 2013 University of Miami
DOI: 10.1111/j.1548-2456.2013.00215.x
Ignacio Arana Araya is a doctoral candidate in political science at the University of Pitts-
burgh. iga3@pitt.edu
Montesinos 2003), and some authors have described the legislators’ effort to influ-
ence the budget (e.g., Siavelis 2002; Baldez and Carey 1999, 2001). However, how
and to what extent informal institutions shape the public budget remains unex-
plored. Some of these studies have aptly suggested that to understand interbranch
relations it is necessary to go beyond the examination of the legal framework and
carefully observe day-to-day political practices. For instance, Ferraro (2008) inter-
views 29 key informants, who reveal how Chilean legislators influence the bureau-
cratic output using informal channels. Ferraro shows that lawmakers influence the
appointment of bureaucratic jobs and build informal networks among public offi-
cials. This study contributes to that line of research, conducting the first systematic
examination of how Congress uses protocols to exert horizontal accountability.1
The absence of studies examining the role of informal institutions in the public
budget has led to a misunderstanding of Chilean executive-legislative relations in
general and the budgetary process in particular. The lacuna is problematic because
the public budget is one of the most important political documents in any country:
it represents the government’s policy choices (Fenno 1966; Wildavsky 1988) and
has a significant impact on electoral outcomes (Ames 1987; Bueno de Mesquita et
al. 2004). When political scientists fail to acknowledge the impact of an informal
institution on a political outcome and attribute its effect to a formal rule, variable
bias is omitted in the causal explanation. As a consequence, the effect of formal rules
tends to be overestimated and our understanding of the phenomenon studied
remains limited.2
The alleged weakness of the Chilean assembly contrasts with the active role
played by other legislatures to increase their budgetary role. Hallerberg et al. (2009)
have documented how budgetary actors in Latin America go beyond their legal
powers and directly contradict the law—or its spirit. This study argues that Chilean
budgetary actors also push the boundaries of their constitutionally limited role, but
in a subtler way: using informal institutions to advance congressional preferences.
Protocols are written agreements signed between the most senior members of
the executive and legislative branches that participate in the budgetary negotiations:
the minister of finance, the president of the congressional Special Budget Joint
Committee (SBJC), and the presidents of the SBJC subcommittees. The protocols
were first created during the negotiations for the 1997 budget. Since then, they have
been signed every year, except in the negotiations for the 2000 budget. The agree-
ments are between 3 and 11 pages long, and contain between 11 and 52 accords that
are usually in effect for one year.3Although protocols are publicly available, they
receive little media attention.4
This article follows the criteria delineated by Helmke and Levitsky (2006) to
argue that protocols are informal institutions. The authors define these institutions as
“socially shared rules, usually unwritten, that are created, communicated, and
enforced outside officially sanctioned channels” (2006, 5). In contrast, they define
formal institutions as “rules and procedures that are created, communicated, and
enforced through channels that are widely accepted as official” (2006, 5). Protocols are
known, respected, and enforced by the political actors that participate in the budgetary
ARANA: CHILEAN BUDGET PROTOCOLS 75
negotiations. This description matches both formal and informal rules. However,
there are two key points that situate protocols “outside officially sanctioned channels.”
First, although protocols are written down, there are no written rules that define
or regulate them. This attribute allows negotiators to reach agreements that are not
limited beforehand. However, it also makes the existence of protocols conditional on
the will of the bargaining actors. Second, unfulfilled protocols do not lead to formal
sanctions. For instance, if the executive does not honor a protocol, it cannot be for-
mally sanctioned. Moreover, if a budgetary actor refuses to sign a protocol, there is
no legal or administrative recourse for reversing the decision or demanding sanctions.
This actually was the situation in 1999, when no protocol was agreed on.
To understand the role of protocols, I conducted semistructured interviews
with 15 key informants involved in the process in Santiago and Valparaíso. As Fer-
raro (2008, 110–11) highlights, the key informant technique has been well devel-
oped in public administration research and has been used by most scholars who have
analyzed the informal relationship between the assembly and the bureaucracy in
Chile (e.g., Montecinos 2003; Siavelis 2002). The main reason is simple yet power-
ful: interviewees provide information that cannot be obtained from other sources.
Since some practices and informal institutions go against the law and are usually
hidden from public view, access to key informants is essential.
This study also conducted a content analysis of the 14 budget protocols that
were agreed on between 1997 and 2011. The systematic examination of the proto-
cols shows that they are a critical bargaining tool that allows Congress to audit the
government and also helps to avoid executive-legislative gridlock. With time, proto-
cols have evolved to become a source of formal institutional change, leading to
changes in relevant laws.
The following section presents evidence that shows the limitations of prevailing
accounts of interbranch relations in Chile. The third section describes how protocols
help the legislature to monitor and oversee the budget, forcing the government to act
with more transparency and to concede to legislative requirements. The fourth sec-
tion argues that protocols are an “accommodating” rule and compares them to other
informal procedures that shape executive-legislative relations in Chile. The conclud-
ing section discusses how protocols shed light on the limits of current rankings of
presidential and legislative powers, the relationship between formal and informal
institutions, and the link between informal rules and democratic outcomes.
THE EXECUTIVE’SBUDGETARY “SUPERPOWERS”:
MYTH AND REALITY
The formal and informal rules that regulate the budgetary bargaining between the
executive and the legislature vary across countries, even within presidential systems.
Scholars view the U.S. federal budget as being jointly determined by presidents and
Congress (e.g., Canes-Wrone 2006; Kiewiet and McCubbins 1991). In contrast,
they claim that in most of Latin America, the power of the purse is strongly biased
toward the executive; constitutions and laws tend to delegate the initiative on
76 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
budget formulation and execution to the government (Hallerberg et al. 2009;
Pelizzo et al. 2005).
Recent studies claim that even in the Latin American context, Chile has one of
the most executive-based models of budget elaboration (Alesina et al. 1999; Vial
2001). Moreover, the supposed main Chilean budgetary actors (i.e., the assembly and
the executive) profess a stringent respect for institutional design (Aninat et al. 2006).
As reported by the OECD, “The current budget process is designed for the Congress
to swiftly approve the government’s proposal with no changes. The influence of the
Congress in informal terms would also appear to be small” (Blöndal and Curristine
2004, 25). Scholars have widely accepted this view, and even one of the most
renowned experts on Chilean politics argues that “it appears that members of Con-
gress are satisfied with a president with strong budgetary powers” (Siavelis 2002, 100).
The budgetary superpowers of the Chilean presidency have helped scholars to
rank it as one of the most powerful in the world. Shugart and Carey (1992) rank the
Chilean presidency as the second most powerful in a sample of 44 countries, while
Samuels and Shugart (2003) place it third among 26 nations. Among works that
examine only Latin America, Shugart and Mainwaring (1997) consider the Chilean
presidency the most powerful, tied with the Colombian; Shugart and Haggard
(2001) classify it in the second position; Tsebelis and Alemán (2005) rank it third;
and Negretto (2008) places it fourth.
Preceding studies have rooted the presidential powers in the 1980 Constitu-
tion, which concentrates almost all the budgetary powers in the executive.5Article
65 of the charter gives the president control over all tax and spending policies, while
article 67 establishes that the executive is responsible for creating a budget bill that
considers all fiscal expenditures. Article 67 also establishes the following restrictions:
• Congress cannot change the economic assumptions or the government’s fore-
casts of future fiscal income.
• If the assembly has not approved the budget bill within 60 days after the exec-
utive’s proposal, it becomes law.
• Congress can reduce the budget only without touching the portion allocated
by previous permanent laws.
• The legislature cannot increase any budget item or reassign expenditures
among programs.6
In sum, the legislature’s participation is formally limited to reducing the flexible
part of the budget, and this must be done within 60 days. The prerogative to reduce
some allocations is of dubious interest to legislators. According to most scholars who
have studied executive-legislative relations in Latin America (e.g., Nacif and Mor-
genstern 2002; Hallerberg et al. 2009; Scartascini et al. 2010) or elsewhere (Poterba
and Von Hagen 1999; Haggard and McCubbins 2000), legislators are mostly inter-
ested in distributing goods and services to their constituencies. Recent research also
suggests that pork distribution is crucial to the electoral survival of Chilean lawmak-
ers (Aninat et al. 2006; Toro 2007). Therefore, representatives have material and
electoral incentives to increase spending. Besides these reasons, members of Con-
ARANA: CHILEAN BUDGET PROTOCOLS 77
gress may be interested in having more influence on the public budget to advance
their ideological preferences and to enhance their role as members of an institution
that oversees the executive.
Thus far, scholars who have discussed the Chilean budgetary process have
assumed that the net effect of the congressional intervention in the public budget is
marginal, and only reduces the executive’s budget bill. However, this assumption is
inconsistent with congressional behavior. Table 1 shows the differences between the
executive’s bill proposal and the budget law that the assembly approved, from 1991
to 2010. It documents that despite restrictions in the charter, the legislature
increased the overall public budget in 7 of the 20 years.
To uncover agency variation, table 2 shows the largest changes among the
budgets of the 22 agencies included in the public budget (19 ministries plus Con-
gress, the judiciary, and the Comptroller General, Contraloría General). It is notable
that in 13 of the years, the largest changes increase the agency budgets up to 13.5
percent. Moreover, in 7 years (1992, 1994, 1997, 1998, 1999, 2007, and 2008), the
allocations of some agencies change while the overall public budget does not (table
1), indicating patterns of reallocation among agencies.
The difference between the amounts in the executive bill and the budget law
results from bargaining practices through which Congress increases the financial
resources of some agencies. Before the protocols are signed, both Congress and the
executive include specifications that change the agencies’ allocations. Specifications
introduce changes that increase or decrease an agency’s budget and can also modify
only parts of its content. While protocols also occasionally reallocate resources, the
legislature uses them mainly to keep the executive accountable.
Tables 1 and 2 challenge the assumptions that have guided preceding research
by documenting that the agencies’ budgetary changes bypass articles 65 and 67 of
78 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
Table 1. Budgetary Variation by Congressional Allocation
Executive Budget Change Executive Budget Change
Bill Law (%) Bill Law (%)
1991 2,509,002 2,514,309 0.2122 2001 10,225,550 10,222,796 –0.9731
1992 3,175,499 3,175,499 0 2002 11,340,114 11,336,360 –0.9669
1993 3,866,243 3,880,877 0.3785 2003 11,889,979 11,890,246 0.0022
1994 4,610,964 4,610,964 0 2004 12,496,288 12,425,588 –0.434
1995 5,505,171 5,474,843 –0.5509 2005 13,102,590 13,102,639 0.0004
1996 6,267,275 6,265,357 0.0306 2006 14,763,933 14,763,971 0.0003
1997 7,135,565 7,135,565 0 2007 17,430,780 17,430,780 0
1998 8,053,023 8,053,023 0 2008 20,209,839 20,209,839 0
1999 8,495,969 8,495,969 0 2009 22,906,583 22,922,583 0.0699
2000 9,274,596 9,274,709 0.0012 2010 25,046,832 25,046,832 0
Note: Numbers are in Chilean million thousand pesos.
Sources: Budget Office (www.dipres.cl) and National Congress (www.congreso.cl).
the constitution. To understand why congressional behavior circumvents the char-
ter, semistructured interviews were conducted with 15 key informants.
The key informants interviewed were selected and contacted on the basis of
their knowledge and direct or indirect participation in the budgetary negotiations.
Among the interviewees were two senators, three deputies, and two former minis-
ters, who span the political spectrum. The other eight interviewees include two ana-
lysts from think tanks associated with political parties and six members of the
Budget Office (Dirección de Presupuestos, which is part of the Finance Ministry); of
these six, three were former directors and the other three were high-level officials.7
All the interviews were conducted between 2004 and 2009.
The interviews uncovered the origin and significance of the protocols: why and
how they emerged, what the signatories pursued with them, why they have evolved,
and what their political implications are. The interviews revealed that Chilean leg-
islators use protocols to advance their preferences in the budgetary process. Their
goal is to increase their influence in the public budget, but they recognize that
attempting to amend the constitution would be too difficult (requiring a majority
of three-fifths of the legislature).
The interviews also revealed that the legislature uses several tactics to raise the
governmental costs of neglecting congressional preferences during the negotiations.
ARANA: CHILEAN BUDGET PROTOCOLS 79
Table 2. Largest Budget Changes for Agencies
Agency Change (%)
1991 Secretariat of the Presidency –2.26
1992 Transport and Telecommunications 5.68
1993 Agriculture 9.22
1994 Secretariat of Government 2.02
1995 Secretariat of Government 13.46
1996 Mining –11.84
1997 Agriculture 2.16
1998 Congress 3.09
1999 Agriculture 5.89
2000 Agriculture 7.94
2001 Congress 2.7
2002 Planning –1.4
2003 Congress 1.81
2004 Congress 1.4
2005 Congress 4.66
2006 Congress 5.61
2007 Housing and Urban Planning –0.6
2008 Secretariat of the Presidency –0.47
2009 Secretariat of Government –1.68
Note: In 2010 there was no variation in ministry budgets.
Sources: Budget Office (www.dipres.cl) and National Congress (www.congreso.cl).
As a consequence of these tactics, the executive usually tries to sidestep a confronta-
tion with Congress and agrees to sign a protocol. Thus, the tactics contribute to the
protocol’s survival.
Among the legislature’s available tactics was one explained by two former
Budget Office directors: members of congress sometimes threaten to reject other
executive bills when the executive does not include their proposed budgetary
amendments. “The budget is not the only bill, so the Finance Ministry has to be
alert for several legal projects in which legislators may retaliate. We have to treat
congressmen extremely well,” said a former Budget Office director in the Eduardo
Frei administration (1994–2000). “We avoid leaving wounds so deep in the nego-
tiations that they may reappear in another bargaining [session],” declared a former
Budget Office director in the Ricardo Lagos government (2000–2006).
Two high officials from the Budget Office and a senator described a second
tactic: lawmakers threaten to escalate the conflict through the media to affect the
government’s popularity. “Sometimes the meetings between the executive and Con-
gress attract the media, giving legislators an opportunity to criticize the executive if
it rejects including sensitive amendments,” commented an SBJC analyst. One SBJC
member and a socialist senator also emphasized this point: “Public discussions are
useful to include changes in the budget.”
Another tactic the legislature uses is delay. Three members of congress stated
that every year, some SBJC members ask their peers to delay the voting of budget
items to gain more time to bargain with the executive. One former SBJC member,
planning minister, and Christian Democratic (centrist) deputy interviewed said that
these petitions usually succeed: “Every legislator may want to offer an amendment,
so legislators respect fair play and support all the petitions.”
Two lawmakers declared that the SBJC takes advantage of the constitutional
rule that only an absolute majority of Congress can approve in the budget law the
fiscal debt beyond the current budgetary year. One legislator said that since Chilean
governments usually incur debts beyond their terms, the representatives threaten to
approve the executive’s bill but reject the debt proposal.
Two SBJC legislators and a former minister noted that Congress has used its
constitutional authority to reduce a budgetary item to a miniscule amount (e.g,
US$2) to force government concessions. A socialist senator recalled in a 2006 inter-
view that the SBJC had used this tactic in 2004, threatening to reduce the allocation
to the Municipal Theater (an opera theater in Santiago) to pressure for an increase
in the money allocated to regional cultural infrastructure. “Only when the govern-
ment accepted our demands did we approve the budget for the Municipal Theater,”
he said. However, the executive can still block this tactic by supplementing expen-
ditures through the Provision for Committed Financing (Organic Law 18,899).8
Although the legislature uses these five tactics to get concessions from the gov-
ernment, protocols are not mere executive “abdications” to Congress. One legislator
and a former Budget Office director interviewed explained that the government has
used the protocols to advance reforms that lack full support within the ruling coali-
tion. According to these informants, sometimes the government makes concessions
80 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
to legislators in exchange for their support for agreements resisted by part of the gov-
erning elite; once the agreements are signed, the executive can claim to have its
hands tied. “In the protocols you frequently see nonbudgetary issues addressed,
especially those related to state management. No human rights issues are included,
but you see agreements about labor policies and corruption,” said one Christian
Democratic deputy interviewed.
PROTOCOLS IN ACTION
For Congress, the budgetary negotiations work against the clock. Every year, on
September 30, the government sends the budget bill directly to the SBJC. The com-
mittee is composed of 13 deputies and 13 senators who proportionately represent
the seats that party coalitions hold in the legislature. The SBJC reviews and amends
the budget bill before sending it to the Chamber of Deputies, which votes on the
proposed bill and sends it to the Senate. The Chamber of Deputies and the Senate
can also amend the bill, but they rarely do so. Thus, the SBJC—controlled by pro-
government forces until the 2010 budget and tied since then—de facto represents
Congress in the budgetary process.
The SBJC has worked year-round since 2003. It is divided into five subcom-
mittees composed of progovernment and opposition legislators. The SBJC has a few
weeks to amend the executive bill before sending it to the Chamber of Deputies,
where it is briefly discussed before sending it to the Senate. The Budget Law must
be approved by November 30 or the executive’s bill becomes law.
Two key informants from the Budget Office said that in the first postauthori-
tarian years there was a tension between the executive and the SBJC members due
to the short timetable to negotiate the budget and the legislators’ frustration with
their limited capacity to influence the process. The informants added that the law-
makers were annoyed with the scant time that they could devote to the negotiations
in electoral years, when elections (held in December) and political campaigns over-
lap with the bargaining period (October and November).9
The increasing interbranch tension reached its limits in the seventh budgetary
negotiations (1996), when the government was having a hard time winning
approval for its bill. To resolve the situation, the deputy Budget Office director,
Mario Marcel, proposed that the SBJC sign an agreement in which the government
would commit to include in the budget relevant congressional demands that were
not incorporated in the bill for the 1997 public budget. The committee agreed.
Unexpectedly, the protocols were created, and have since become a relevant tool
that is used to close the annual negotiations, after the SBJC votes amendments to
the executive’s bill.10
With the first six-page protocol signed on November 20, 1996, the budget law
for 1997 was promulgated, and the government started to share parts of its exclusive
budgetary and legislative powers with the assembly. Interviewed in 2006, Marcel, by
then a former Budget Office director, strongly valued his creation: “Protocols allow
agreeing to things that cannot be solved in the regular negotiations. They have been
ARANA: CHILEAN BUDGET PROTOCOLS 81
a great contribution to the budgetary discussion and have strengthened the collab-
oration between the executive and the SBJC.”
Progovernment and opposition SBJC legislators interviewed also valued the
function that protocols serve and expressed a strong commitment to maintaining
them. One SBJC member and senator for the Independent Democratic Union
(rightist) said that the protocols are “the most useful tool to oversee the executive in
the budgetary process.” Another SBJC member and Christian Democratic senator
argued that protocols “are very effective for reducing the tension between govern-
ment and Congress, and to avoid focusing all the negotiations on the budget law.
Protocols establish a timetable of issues that are resolved during the year with the
government’s commitment.”
Strict enforcement explains why protocols are a reliable mechanism for aiding
negotiations. Two legislators stated that if the accords were not executed, the gov-
ernment would risk a conflict with the legislature and a loss of credibility among
voters. Since current agreements rely on the successful implementation of previous
ones, unfulfilled protocols would probably decrease the life expectancy of the infor-
mal rule. However, both the government and Congress have vested interests in sign-
ing the protocols.
To ensure that protocols are enforced, the two elected branches of power
closely monitor the agreements. Two lawmakers interviewed mentioned that in case
of noncompliance, the SBJC can demand explanations directly from a responsible
minister. However, when asked, none of the SBJC legislators interviewed reported
a serious unfulfilled promise.
In the first five introductory paragraphs of the first protocol, the signatories
(including the finance minister and the presidents of the SBJC and its subcommit-
tees) argued that their work was intended to build mechanisms that stimulated the
efficiency in the management of public institutions, to reinforce transparency in the
administration and allocation of public resources, and to dispatch the bill according
to constitutional deadlines. These three reasons were recognized in subsequent pro-
tocols, in which the signatories added, as motivations for the agreements, the exec-
utive’s commitment to recognize and execute the spirit of amendments proposed by
the legislators but legally inadmissible because they entailed issues of exclusive pres-
idential initiative (Dipres 1997, 1998), “several issues” that needed to be assessed
but were not included in the budget law (Dipres 1998), the intention to improve
the procedures for the budgetary discussion, the need to emphasize employment
creation in the budget, the need to develop studies to enrich future budgetary dis-
cussions (Dipres 2000), and the intention to strengthen the government’s commit-
ment to fiscal discipline (Dipres 2001).
Among these nine reasons, all but the third (meeting constitutional deadlines)
suggest the same overall goal: to complement the executive’s bill with measures that
lead to improved accountability to Congress of the budgetary process. Moreover,
the fourth reason (attention to extra amendments) illustrates an explicit agreement
to bypass the constitution, because the government commits to include in the
budget law what the charter prohibits to the assembly. This is unambiguous in the
82 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
protocol for the 1998 budget (Dipres 1997, 1), which recognizes that “several
deputies and senators presented amendments that were declared inadmissible
because they involve matters of exclusive presidential initiative…. This protocol
expresses the executive’s commitment to implement those aspects raised in the spirit
of the amendments.”
The protocols cannot be quantitatively estimated because their content is
mainly qualitative. From the 389 agreements contained in the protocols between
1997 and 2011, the government committed to action in 387 and Congress in 15. I
classified the accords contained in the protocols in eight categories: those that pro-
mote information disclosures; entail a governmental concession; commission
research; request the evaluation of programs; increase, limit, and reallocate expendi-
tures; and commit the government to send a bill.11 Figure 1 summarizes the content
of the agreements according to their type.
The figure shows that protocols enhance horizontal accountability. Following
Schedler (1999), accountability entails answerability and enforcement. Answerabil-
ity requires officeholders to report and explain what they have done or plan to do.
Enforcement is the capacity of accounting organizations to sanction officeholders
who have violated their duties. Schedler (1999, 14) proposes that the concept of
accountability implies controlling those in power through the threat of sanctions,
forcing them to justify their acts and to act with transparency. Thus, exercising
accountability involves monitoring, oversight, and enforcement capacity. That is
precisely what protocols do.
The first four categories described account for 83 percent of the agreements,
and they are all focused on making the government more accountable. Information
disclosure, the executive concessions, the commission of research, and evaluations
ARANA: CHILEAN BUDGET PROTOCOLS 83
Figure 1. Agreements by Type, 1997–2011
Source: Protocols for the 1997–2011 public budgets.
increase the budgetary information released to Congress and make the budgetary
process more transparent. These categories show that protocols shed light into the
black box of the public budget, obligating the government to release information
and justify the continuity of programs, while Congress increases its monitoring and
oversight functions.
Each type of agreement generates interesting patterns of executive-legislative
negotiations. At a minimum, the agreements through which the government com-
mits to release information lead to a more transparent allocation of resources and
enhance the congressional capacity to oversee. This type of accord is evident in the
text of the documents. For instance, in the protocol for the 1997 budget (Dipres
1996, 2), the government agrees to include in the budget law a norm mandating
that public services that transfer resources to foundations receive in exchange “an
annual financial balance, a list of programs and activities conducted, and a list of
their board members and senior executives.” The disclosure of this kind of infor-
mation allows lawmakers to detect the political motivations behind support for
foundations.
The executive’s concessions to Congress are an umbrella category whereby the
government commits to modify a public policy or introduce change in the planning
or execution of the budget that cannot be classified in other categories. These con-
cessions allow the assembly to influence the budgetary process. The concessions can
be written in general or specific terms. In an example of the latter, the protocol for
the 2001 budget (Dipres 2000, 2) states that “to update medium-term financial
forecasts, reports about bills with fiscal impact will cover four years.”
The commissioning of research forces the government to justify its decisions
and also empowers representatives to promote their policy agenda. For instance, in
the protocols for the 2004 budget (Dipres 2003, 5), the government agrees to com-
mission studies of “alternatives to the expansion of preschool coverage and support
for the economic integration of women” and “policies for the farthest regions.” The
expansion of welfare policies, the role of women in the labor force, and plans for
regional development allow lawmakers to promote these causes, distribute pork, and
claim credit for it.
The evaluation of programs allows legislators to assess a public policy that they
(or the executive) want to change or replace. For instance, in the protocol for the
2007 budget (Dipres 2006, 9), the government agreed to assess the Housing Assis-
tance Program of the Ministry of Housing and Urban Planning “to adjust such pro-
gram to suit the objectives and instruments of the new housing policy.” Such an
evaluation is likely to be rejected by civil servants who work in the program or politi-
cians who may benefit from it. Its inclusion in the protocol nonetheless makes the
evaluation unavoidable.
In addition to the evaluations described in the protocols, since 2001 the gov-
ernment has committed to thoroughly assessing programs that are listed in the
appendixes of the protocols. As shown in figure 2, these evaluations have increased
over time. Evaluations increased from an average of 17.8 for 2001–6 to 33.4 for
2007–11.
84 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
The agreements that discuss limits, increases, or reallocations of expenditures
play a minor role in the protocols; they represent only 9 percent of the agreements.
This is mainly because the amounts for the next fiscal year are previously negotiated.
Protocols tend to describe expenditures broadly, committing the executive to
allocate (or not) resources in a certain direction. For instance, the protocol for the
2011 budget (Dipres 2010, 4–5) states that “the government commits to incorpo-
rate more resources to the Labor Direction, to the extent that there is available
budget.” In the same vein, the protocol for the 2002 budget (Dipres 2001, 1) states
that “during 2002 the spending on honorariums on public services cannot be
increased through transfers from the Treasury.”
Figure 3 shows that the number of agreements contained in the protocols has
been increasing over time: there were 210 in 2007–11 and only 179 in the nine pre-
vious years. This trend suggests that protocols have gained more relevance in recent
years. The two elected branches of power could be interested in signing more
accords; protocols serve the SBJC by including more of its preferences in the budget,
while the government gets more legislative support for reforms.
Although protocols were created to extend congressional influence in the
budget process and improve control over the executive, with time the agreements
have evolved to include issues beyond the budget. This is particularly clear in the
accords, in which the government agrees to send bills. In them, the executive not
only concedes to sharing its legislative powers; many of the bills refer to nonbud-
getary issues, and indeed, some of them involve extremely relevant political issues.
This seeming “abdication” by the executive could suggest that the Chilean govern-
ment underuses its power, but it may actually work in the government’s interest. As
key informants have described, the government sometimes uses the protocols to
advance reforms that are difficult to achieve otherwise due to intracoalition con-
ARANA: CHILEAN BUDGET PROTOCOLS 85
Figure 2. Evaluations over Time
Source: Protocols for the 1997–2011 public budgets.
flicts. Table 3 summarizes the content of the bills that the executive has agreed to
send to Congress.
As the table shows, the executive started committing to send bills in 2001 and
the number increased after 2003, suggesting a growing rate of shared legislative
responsibilities. Some promised bills even refer to changes in organic constitutional
laws. These laws complement the constitution in certain subjects and regulate
highly important issues and institutions, such as the public educational system, the
central bank, Congress, the national police, the armed forces, municipalities, polit-
ical parties, the constitutional court, and the Comptroller General. An organic con-
stitutional law can be changed only with a very high quorum: four-sevenths of the
active members of Congress.
The fact that the government and the Congress use the protocols to change laws
related to the constitution demonstrates how an informal institution can modify formal
rules. In this case, a byproduct of the constitution (the protocols) can end up modifying
laws that complement one of the most sacred formal institutions a country has.
Moreover, some of the bills that the executive has submitted in response to a
protocol have had a direct impact on the congressional capacity to oversee the very
executive that proposed them. For instance, the SBJC acquired permanent status
after an agreement signed in the protocol for the 2003 budget (Dipres 2002, 1).
This change of status led to a virtuous cycle of horizontal accountability: due to a
reform to article 19 of Law 18,918, since 2003 the SBJC meets more frequently and
has more time and financial resources to discuss and oversee the public budget.
Table 3 also shows that for three consecutive years (2007, 2008, and 2009), the
executive committed to send a bill to reform the organic constitutional law that regu-
lates the Comptroller General. This bureau was reformed after the second attempt,
but negotiators agreed on further reforms. The Comptroller General’s office is an
86 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
Figure 3. Agreements per Year
Source: Protocols for the 1997–2011 public budgets.
independent institution whose main goal is to put a check on the government: it has
a constitutional mandate to review and approve the legality of all documents issued by
the state. The reforms were intended to change an institution that scholars and politi-
cians had widely depicted as outdated and unable to oversee the executive (e.g., Waiss-
bluth 2005; Harasic 2006). “The Contraloría has too few powers, a small staff, and its
work is not well focused,” said a senator for the Independent Democratic Union.
The changes to the Comptroller General enhanced horizontal accountability,
increasing the capacity of an independent agency to audit the government. In the
same vein, the government also committed to sending other bills that have served as
a platform to increase horizontal accountability. Laws such as the 2006 rule on fiscal
responsibility (Law 20,128) and the 2008 law on transparency (Law 20,285) forced
the government to release more information, justify its actions, and follow proce-
dures that constrain its behavior.
ARANA: CHILEAN BUDGET PROTOCOLS 87
Table 3. Executive Commitments, 2001–2011
2001 Regulation of civil servants’ rights 2008 Reform of corporate governance of
and duties public companies
2003 Strengthening of regional governments 2008 Creation of a long-term investment
budget for the armed forces
2003 Acceleration of administrative 2008 Reform of the Comptroller General
investigations
2003 Standardization of the adjudication 2009 Reform of the Chilean prison guard
of public investment contracts force
2003 New rules related to state financial and 2009 Create pension program for public
personnel administration university staff
2003 Grant permanent status to the SBJC 2009 Reform of the Comptroller General
2005 Creation of the Differentiated 2009 Creation of the bicentennial
Educational Grant scholarship fund
2005 Improvement of fiscal responsibility 2010 Reform of the retirement bonus law
and transparency
2007 Replacement of the labor bonus for 2011 Creation of an ethical family income
remote regions standard
2007 Creation of the Education 2011 Regulation of mine closures
Superintendent
2007 Reform of the National Sports Institute 2011 Reform of the copper reserve law
2007 Expansion of the senior public 2011 Redesign of graduate scholarships
management system
2007 Reform of the employment contract 2011 Reform of the financial management
for civil servants of public universities
2007 Reform of the Comptroller General 2011 Reform of the senior public
management system
2008 Enhancement of remote region
development
Source: Protocols for the 1997–2011 public budgets.
PROTOCOLS AS AN INFORMAL RULE
Helmke and Levitsky (2006) classify the informal institutions in four categories
according to their interaction with formal rules: complementary, accommodating,
substitutive, and competing. These categories are defined in relation to two dimen-
sions. One is the degree of convergence between formal and informal institutional
outcomes. If the outcome is substantively different, then formal and informal insti-
tutions diverge. Otherwise, they converge. The second dimension relates to the
effectiveness of the formal institutions, something directly correlated with the prob-
ability of enforcement.
Protocols are an accommodating informal institution. This type of rule com-
bines effective formal institutions and divergent outcomes (Helmke and Levitsky
2006, 15). It is often created by actors who dislike the effects generated by the
formal rules but cannot change them. The Chilean legislators dislike their budgetary
role, and although they could attempt to reform the charter to change the status
quo, it is a very difficult task. As Elkins et al. (2009, 67) remark, any constitutional
reform is costly, because bargaining actors need to invest significant time and polit-
ical capital until the final outcome takes place. Attempting to amend the 1980 char-
ter is particularly costly because it requires a very high quorum (three-fifths of the
congress).
The protocols’ dynamic capacity to change formal rules also aligns with
Helmke and Levitsky’s expectation. These authors claim that due to lower coordi-
nation costs, informal institutions that are the product of the interaction of a small
number of actors are more sensitive to change than rules that involve more actors
(2006, 22). Protocols are created by a small group; only 26 SBJC legislators, the
finance minister, and a few members of the Budget Office bargain them.
The small size and low turnover of the group that negotiates the public budget
also contribute to the protocols’ endurance. The SBJC members tend to stay on the
committee during the four-year legislative term, and several of them have served on
it for more than one period. The executive’s negotiators have also been stable:
throughout the five postauthoritarian governments, there have been only six finance
ministers and six Budget Office directors. Since the budgetary negotiators are spe-
cialists in the public budget and know each other, they have accumulated interper-
sonal trust, and can also anticipate the bargaining strategy of other actors. Therefore,
all members of the bargaining group know how to negotiate protocols and what
they can get from them. Furthermore, the “shadow of the future”—the possibility
that the bargaining players can suffer retaliation if they do not cooperate—also helps
to reduce transaction costs.
Other authors have also described elite-created informal practices in Chilean pol-
itics. Siavelis (2006) argues that the political elite have used accommodating rules such
as the cuoteo, a formula for allotting government positions and parliamentary candi-
dacies; the partido transversal, a supraparty core of elites within the governing coali-
tion; and a pattern of informal agreements known as democracia de los acuerdos, to
reach agreements. Carey and Siavelis (2006) describe as a complementary institution
88 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
the “electoral insurance,” which allowed the ruling center-left Concertación coalition
to compensate high-profile politicians who competed for congressional seats that were
difficult to win. Of these four institutions, the partido transversal and the democracia
de los acuerdos do not exist anymore, and it is not yet clear whether the “electoral insur-
ance” has survived the defeat of the Concertación in 2010.12 The cuoteo, then, is the
only informal rule that has clearly survived the 2010 transfer of power.
The democracia de los acuerdos and the partido transversal were highly sensitive
to the political environment and relied on an unidentified group of influential
politicians. Perhaps their dependence on such feeble bases condemned them to dis-
appear. In contrast, the survival of the cuoteo, the “electoral insurance,” and the pro-
tocols depends on formal institutions. The first two rest on the electoral system,
whose continuity is uncertain for the long term due to the continuous attempts of
the Concertación to change it. The protocols depend on a seemingly more stable
formal institution; none of the political coalitions have attempted to change articles
65 and 67 of the constitution, despite several reforms to the charter since 1980.
A relevant difference between protocols and the other informal rules mentioned
is that the former are the only written agreements. This and the fact that protocols
are publicly available contribute to their enforcement because they force the govern-
ment to honor its word. Another difference is that protocols are the only informal
rule that has evolved to become a source of change of formal institutions. Moreover,
protocols are the only informal institution that violates the constitution—in fact,
three articles of it. Some agreements signed in the protocols infringe on article 65
because they force the executive to share its spending policies, and on article 67
because they allow Congress to increase and reallocate expenditures. The agreements
also transgress article 53, which prohibits the Senate from overseeing the acts of the
government and all the entities under its dependency or adopting agreements that
imply overseeing.
CONCLUDING REMARKS
This article has shown that protocols are a key tool that contributes to reaching
interbranch agreements. Protocols help the Chilean Congress to monitor and over-
see the government in the budgetary process, forcing the executive to act with more
transparency and to concede to legislative requirements. Protocols also permit the
government to advance sensitive reforms that are resisted within the ruling coalition.
This informal institution contributes to avoiding executive-legislative gridlock.
The analysis demonstrates that protocols challenge previous accounts of budg-
etary negotiations in Chile. It also defies the depiction of the budgetary actors as
outliers in Latin America due to their (allegedly) rigorous adherence to law.
This study has argued that protocols are an informal institution because they
are not regulated by any written rule and the negotiators cannot be formally sanc-
tioned for not complying with them. These characteristics do not mean that proto-
cols are symbolic; as textual analysis and interviews with key informants reveal, the
agreements are vigorously enforced.
ARANA: CHILEAN BUDGET PROTOCOLS 89
This analysis of the protocols suggests broader implications. First, the ability of
the protocols to check the power of the executive should make us reconsider presi-
dential power rankings—or at least take them with a grain of salt. The classification
of presidential and legislative powers should be carefully reexamined to avoid over-
or underestimating the effect of formal institutions on political outcomes by omit-
ting the impact of informal rules. A systematic cross-national comparison of how
informal procedures mitigate presidential powers is probably unfeasible due to the
idiosyncratic nature of informal institutions. However, an index of constitutional
enforceability can take into account the limits of formal rules.
Second, the protocols are an example of how an informal institution can
become a source of formal institutional change: new laws, institutions, and policies
have been created under the flexible framework that protocols provide. This charac-
teristic suggests that scholars should also pay attention to how formal and informal
rules interact and shape each other over time, because this dynamic relationship can
produce dissimilar outcomes at different points in time.
Third, this study has shown how informal institutions may lead to positive
democratic outcomes. Besides enhancing horizontal accountability, protocols pro-
mote the settlement of executive-legislative disputes, and thus can be depicted as
reinforcing governability. This approach is scant in the literature, which has con-
tributed to sustaining the longstanding image of Latin American presidentialism as
characterized by executive dominance unconstrained by institutional checks (e.g.,
Boix 2003). Previous studies have associated informal institutions with nepotism,
corporatism, clientelism, and patrimonialism. For instance, Hartlyn (1998) shows
that norms of patrimonialism lead to an executive dominance that goes far beyond
constitutional norms. As noted, protocols work exactly in the other direction.
Despite the positive effects of protocols, they are not the ideal means to exert
horizontal accountability. In Chile there is a popular figure known as the maestro
chasquilla (akin to a handyman), usually a low-cost worker who performs all kinds
of domestic services. He usually provides temporary and creative solutions for cus-
tomers who want to reduce costs. However, his work often is associated with indi-
rect and long-term costs (e.g., a defective repair leaves a permanent leak). Like the
repairs offered by a maestro chasquilla, protocols are a cheap, second-best, short-term
solution that may lead to indirect and long-term costs.
Protocols are a second-best solution precisely because they are a short-term remedy
whose endurance depends on annual negotiations. A constitutional amendment would
provide a permanent solution, but that option would entail a herculean legislative
effort; agreeing on a protocol is cheaper. The indirect and long-term cost of protocols
is that they discourage efforts to work toward a permanent reform that could empower
the assembly. This informal institution, which has no legal standing and which receives
little media coverage, does allow a volatile congressional oversight. In contrast, a legis-
lature with a formally stronger budgetary role would force members of congress to be
publicly responsible for overseeing the budget. Furthermore, a constitutional reform
would increase the incentives for constituencies and civic organizations to put pressure
on an empowered Congress to have their preferences represented in the budget.13
90 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
Despite prominent contributions (e.g, Helmke and Levitsky 2006; Hallerberg et
al. 2009), we still know very little about the influence of informal institutions in Latin
American politics. To explore informal rules is difficult and time-consuming: the
researcher often needs to do fieldwork to uncover their existence and examine data
that can be difficult to gather, measure, and test. However, if political scientists want
to understand better how politics actually works, this is a task that cannot be ignored.
NOTES
I am very grateful to Aníbal Pérez-Liñán, Scott Morgenstern, Brian Phillips, Miguel
Carreras, and three anonymous reviewers for LAPS for their thoughtful comments and useful
suggestions to improve this article.
1. Montecinos (2003, 19) was the first scholar to notice that protocols emerged due
to “congressional demands for greater accountability” in the public budget. However, she did
not methodically analyze them, perhaps because of the unpredictable future of the few agree-
ments signed at the time she conducted her research.
2. The bias can also lead to an underestimation of formal rules. This happens when
the unestimated impact of informal institutions on political outcomes contradicts the effect
of formal institutions.
3. Since protocols usually have an annual duration, some of the agreements are rene-
gotiated over the years. That explains why 14 percent of the 389 agreements made between
1997 and 2011 are repeated in the protocols.
4. Most protocols can be found on the webpage of the Budget Office (www.dipres.cl).
The only exceptions are the agreements bargained for the 1997 and 1998 budgets, which
were requested from the Finance Ministry for this work. Protocols do not get much media
coverage: the data collection process used the electronic database of the main Chilean news-
paper, El Mercurio, to examine all news that mentioned the protocols in the most influential
media. Little useful information was found.
5. Constitutional designers reinforced the executive to legitimate the concentration of
powers that characterized the Pinochet dictatorship and to prevent a conflict with Congress
once elections were reinstalled.
6. In addition to these constraints, Congress cannot oversee the purchase of weapons
by the military based on the income the military receives from the Corporación del Cobre
(Codelco), the copper state company.
7. All the key informants interviewed were asked about the budgetary negotiations,
their personal role and goals in them, the relevance of the protocols, and their assessment of
all their counterparts in the negotiations. While politicians described how the negotiations fit
into the broader scenario of interbranch relations, informants from the Budget Office and the
think tanks provided a more technical view of the budgetary actors and process.
8. The Provision for Committed Financing is a tool that allows the executive to
finance unexpected contingencies and new budgetary requirements.
9. In fact, the press has noted that legislators approve the budget law faster in electoral
years to concentrate their energy on the campaigns (e.g., El Mercurio 2000, 2005).
10. Protocols are the last step before Congress approves the budget law. Among the 14
protocols analyzed, 13 were signed between November 7 and November 25, just before the
congressional approval deadline (November 30). An exception was the protocol for the 1999
budget, signed after Congress approved the law, on December 16.
ARANA: CHILEAN BUDGET PROTOCOLS 91
11. The negotiators can also include in the protocols promises regarding future public
budgets, especially if the accord cannot be implemented in the budget under negotiation. For
instance, in the protocol for the 1999 budget (Dipres 1998, 5), the executive committed to
reallocate resources within the Defense Ministry in the 2000 budget.
12. The next congressional elections will be held in 2013.
13. One alternative view could propose that if protocols become increasingly relevant
and lead to a de facto constitutional change, they may pave the way to amend the charter. It
could also be proposed that the charter be updated to avoid the current transgressions to arti-
cles 65, 67, and 53. However, a constitutional reform has never been mentioned in the pro-
tocols, and it did not show up as a probable event in the interviews conducted for this study.
REFERENCES
Alesina, Alberto, Ricardo Hausmann, Rudolph Hommes, and Ernesto Stein. 1999. Budget
Institutions and Fiscal Performance in Latin America. Journal of Development Economics
59, 2 (August): 253–73.
Ames, Barry. 1987. Political Survival: Politicians and Public Policy in Latin America. Berkeley:
University of California Press.
Aninat, Cristóbal, John Londregan, Patricio Navia, and Joaquín Vial. 2006. Political Institu-
tions, Policymaking Processes, and Policy Outcomes in Chile. Unpublished mss. Wash-
ington, DC: Inter-American Development Bank.
Baldez, Lisa, and John M. Carey. 1999. Presidential Agenda Control and Spending Policy:
Lessons from General Pinochet’s Constitution. American Journal of Political Science 43,
1 (January): 29–55.
———. 2001. Budget Procedure and Fiscal Restraint in Posttransition Chile. In Haggard
and McCubbins 2001. 105–48.
Blöndal, Jón R., and Teresa Curristine. 2004. Budgeting in Chile. OECD Journal of Budget-
ing 4, 2: 7–45.
Boix, Carles. 2003. Democracy and Redistribution. Cambridge: Cambridge University Press.
Bueno de Mesquita, Bruce, Alastair Smith, Randolph M. Siverson, and James D. Morrow.
2004. The Logic of Political Survival. Cambridge: MIT Press.
Canes-Wrone, Brandice. 2006. Who Leads Whom? Presidents, Policy, and the Public. Chicago:
University of Chicago Press.
Carey, John M., and Peter Siavelis. 2006. Election Insurance and Coalition Survival: Formal
and Informal Institutions in Chile. In Helmke and Levitsky 2006. 160–77.
Dirección de Presupuestos (Dipres). 1996. Protocolo de Acuerdo de la Ley de Presupuestos
de 1997. Santiago, Chile.
———. 1997. Protocolo de Acuerdo de la Ley de Presupuestos de 1998. Santiago, Chile.
———. 1998. Protocolo de Acuerdo de la Ley de Presupuestos de 1999. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-42512.html> Accessed December 8, 2010.
———. 2000. Protocolo de Acuerdo de la Ley de Presupuestos de 2001. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-42511.html> Accessed December 8, 2010.
———. 2001. Protocolo de Acuerdo de la Ley de Presupuestos de 2002. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37471.html> Accessed December 8, 2010.
———. 2002. Protocolo de Acuerdo de la Ley de Presupuestos de 2003. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37466.html> Accessed December 8, 2010.
———. 2003. Protocolo de Acuerdo de la Ley de Presupuestos de 2004. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37462.html> Accessed December 8, 2010.
92 LATIN AMERICAN POLITICS AND SOCIETY 55: 4
———. 2004. Protocolo de Acuerdo de la Ley de Presupuestos de 2005. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37458.html> Accessed December 8, 2010.
———. 2005. Protocolo de Acuerdo de la Ley de Presupuestos de 2006. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37451.html> Accessed December 8, 2010.
———. 2006. Protocolo de Acuerdo de la Ley de Presupuestos de 2007. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37443.html> Accessed December 8, 2010.
———. 2007. Protocolo de Acuerdo de la Ley de Presupuestos de 2008. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-37003.html> Accessed December 8, 2010.
———. 2008. Protocolo de Acuerdo de la Ley de Presupuestos de 2009. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-42525.html> Accessed December 8, 2010.
———. 2009. Protocolo de Acuerdo de la Ley de Presupuestos de 2010. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-54796.html> Accessed December 8, 2010.
———. 2010. Protocolo de Acuerdo de la Ley de Presupuestos de 2011. Santiago, Chile.
<http://www.dipres.gob.cl/572/article-72443.html> Accessed December 8, 2010.
Elkins, Zachary, Tom Ginsburg, and James Melton. 2009. The Endurance of National Con-
stitutions. New York: Cambridge University Press.
Fenno, Richard. F., Jr. 1966. The Power of the Purse: Appropriations Politics in Congress.
Boston: Little, Brown.
Ferraro, Agustín. 2008. Friends in High Places: Congressional Influence on the Bureaucracy
in Chile. Latin American Politics and Society 50, 2 (Summer): 101–29.
Haggard, Stephan, and Matthew D. McCubbins. 2001. Presidents, Parliaments, and Policy.
New York: Cambridge University Press.
Hallerberg, Mark, Carlos Scartascini, and Ernesto Stein. 2009. Who Decides the Budget? A
Political Economy Analysis of the Budget Process in Latin America. Cambridge: Harvard
University Press.
Harasic, Davor. 2006. Fin a “secretismos” y silencio. El Mercurio, August 12.
Hartlyn, Jonathan. 1998. The Struggle for Democratic Politics in the Dominican Republic.
Chapel Hill: University of North Carolina Press.
Helmke, Gretchen, and Steven Levitsky. 2006. Informal Institutions and Democracy: Lessons
from Latin America. Baltimore: John Hopkins University Press.
Kiewiet, D. Roderick, and Matthew D. McCubbins. 1991. The Logic of Delegation: Congres-
sional Parties and the Appropriations Process. Chicago: University of Chicago Press.
Mainwaring, Scott, and Christopher Welna. 2003. Democratic Accountability in Latin Amer-
ica. New York: Oxford University Press.
Marcel, Mario. 2006. Former Director, Budget Office. Author interview. Santiago, March 18.
El Mercurio (Santiago). 2000. Presupuesto aprobado en plazo récord. November 16.
———. 2005. Sesión falla por inasistencia parlamentaria. September 16.
Montecinos, Verónica. 2003. Economic Policy Making and Parliamentary Accountability in
Chile. UNRISD Democracy, Governance and Human Rights series, no. 11. Geneva:
United Nations Research Institute for Social Development.
Morgenstern, Scott, and Benito Nacif, eds. 2002. Legislative Politics in Latin America. New
York: Cambridge University Press.
Negretto, Gabriel. 2008. Political Parties and Institutional Design: Explaining Constitutional
Choice in Latin America. British Journal of Political Science 39, 1: 117–39.
O’Donnell, Guillermo. 1998. Horizontal Accountability in New Democracies. Journal of
Democracy 9, 3 (July): 112–26.
Pelizzo, Riccardo, Rick Stapenhurst, and David Olson. 2005. The Role of Parliaments in the
Budget Process. Washington, DC: World Bank.
ARANA: CHILEAN BUDGET PROTOCOLS 93
Poterba, James M., and Jürgen von Hagen. 1999. Fiscal Institutions and Fiscal Performance.
Chicago: University of Chicago Press.
Samuels, David J., and Matthew Soberg Shugart. 2003. Presidentialism, Elections, and Rep-
resentation. Journal of Theoretical Politics 15, 1 (January): 33–60.
Scartascini, Carlos, Ernesto Stein, and Mariano Tommasi. 2010. How Democracy Works:
Political Institutions, Actors, and Arenas in Latin American Policymaking. Washington,
DC: Inter-American Development Bank.
Schedler, Andreas. 1999. Conceptualizing Accountability. In The Self-Restraining State:
Power and Accountability in New Democracies, ed. Schedler, Larry Diamond, and Marc
Plattner. Boulder: Lynne Rienner. 13–28.
Shugart, Matthew Soberg, and John M. Carey. 1992. Presidents and Assemblies: Constitutional
Design and Electoral Dynamics. New York: Cambridge University Press.
Shugart, Matthew Soberg, and Scott Mainwaring. 1997. Presidentialism and Democracy in
Latin America: Rethinking the Terms of the Debate. In Presidentialism and Democracy
in Latin America, ed. Mainwaring and Shugart. New York: Cambridge University Press.
12–54.
Shugart, Matthew Soberg, and Stephen Haggard. 2001. Institutions and Public Policy in
Presidential Systems. In Haggard and McCubbins 2001. New York: Cambridge Uni-
versity Press. 64–104.
Siavelis, Peter. 2002. Exaggerated Presidentialism and Moderate Presidents: Executive-Leg-
islative Relations in Chile. In Morgenstern and Nacif 2002. 7–13.
———. 2006. Accommodating Informal Institutions and Chilean Democracy. In Helmke
and Levitsky 2006. 33–55.
Toro, Sergio. 2007. Conducta legislativa ante las iniciativas del ejecutivo: unidad de los blo-
ques políticos en Chile. Revista de Ciencia Política 27, 1: 23–41.
Tsebelis, George, and Eduardo Alemán. 2005. Presidential Conditional Agenda Setting in
Latin America. World Politics 57, 3 (April): 396–420.
Vial, Joaquín. 2001. Institucionalidad y desempeño fiscal: una mirada a la experiencia de
Chile en los 90. Serie Estudios Socio/Económicos 5: 1–39.
Waissbluth, Mario. 2006. La reforma del estado en Chile 1990–2005, de la confrontación al
consenso. Barcelona: ESADE.
Wildavsky, Aaron B. 1988. The New Politics of the Budgetary Process. Glenview: Scott, Fores-
man.
94 LATIN AMERICAN POLITICS AND SOCIETY 55: 4