Mar 26, 2015
Motivation• Economists, IFIs emphasize policy recipes to achieve
development goals.• This led to adoption of “Washington Consensus” reforms.• Outcome somewhat disappointing.• IPES: Potential of policy recipes depends on the quality of the
policymaking process through which policies are discussed, approved, and implemented.
• Improvement and better understanding of policymaking processes key in order to improve the quality of public policies.
Beyond technocratic approach to policymaking• In technocratic approach, public policies are objects of choice. Policymakers just
need to adopt those that lead to best outcomes.• Actual policymaking involves a multiplicity of political actors interacting in a
variety of arenas.• Policies are outcome of inter-temporal transactions among them.• IPES: less emphasis on content of policies, more on process of policymaking
(PMP), and some resulting policy features.• These processes differ across countries, can be traced back to each country’s
political institutions.• While institutions are also endogenous, depend on culture and historical legacies,
in IPES we treat them mostly as exogenous.– Focus on impact of existing institutions on policy outcomes, not on the dynamics of
institutional change.
Political institutions and policy outcomes• So how do political institutions affect policy outcomes? • Literature tends to focus on single institutional dimensions as determinants
of some policy outcome.• PMPs do not depend on single-factor explanations but rather on a
multiplicity of factors, and their interaction.• We adopt systemic approach, more appropriate to:
– understand why policy reforms that work under certain institutional environments may not work under others
– think about the reform of political institutions.
• Approach requires deep knowledge of institutional details of the countries under study.
Country focus• Phase I of research agenda focused on detailed country
studies, produced by local teams of economists and political scientists.
• Argentina, Brazil, Chile, Colombia, Ecuador, Mexico, Paraguay, Peru, Uruguay and Venezuela.
• We have added Costa Rica, Guatemala and Jamaica.
• Country studies are key input for the report.
Key features of public policies• The focus of the approach is in explaining certain
common features of public policies:– Stability.
– Adaptability.
– Coordination and coherence.
– Quality of implementation and enforcement.
– Public regardedness.
– Efficiency.
Measuring key features of policies• Two main sources of data
• International comparative indicators from the Global Competitiveness Report (GCR).
• Our own State Capabilities Survey, responded by more than 150 experts in 18 countries in Latin America.
• Time frame: last two decades (or since return of democracy).
• Thus, our indices do not reflect quality of policymaking under current administrations, but during this longer period.
Example: Measuring policy stability• Volatility of Fraser index of Economic Freedom.
• GCR question regarding cost of legal and political changes for firm’s planning capacity.
• GCR question on the extent to which new governments honor contractual commitments and obligations of previous regimes.
• SC Survey question on capacity of the State to set and maintain policy priorities among conflicting objectives.
• SC Survey question on extent to which governments ensure policy stability so that policies have time to work.
• SC Survey question on extent to which the State makes and maintains international commitments.
Example: Measuring policy stability• Volatility of Fraser index of Economic Freedom.
• GCR question regarding cost of legal and political changes for firm’s planning capacity.
• GCR question on the extent to which new governments honor contractual commitments and obligations of previous regimes.
• SC Survey question on capacity of the State to set and maintain policy priorities among conflicting objectives.
• SC Survey question on extent to which governments ensure policy stability so that policies have time to work.
• SC Survey question on extent to which the State makes and maintains international commitments.
1 2 3 4
Venezuela
Argentina
Ecuador
Guatemala
Nicaragua
Bolivia
Paraguay
Dominican Rep.
Peru
Panama
Honduras
El Salvador
Mexico
Colombia
Brazil
Costa Rica
Uruguay
Chile
Stability Index (1-4 scale)
Source: Authors’ compilations.
Policy Stability index
1 2 3 4
Venezuela
Argentina
Ecuador
Guatemala
Nicaragua
Bolivia
Paraguay
Dominican Rep.
Peru
Panama
Honduras
El Salvador
Mexico
Colombia
Brazil
Costa Rica
Uruguay
Chile
Stability Index (1-4 scale)
Source: Authors’ compilations.
Policy Stability index
HighHigh
LowLow
MediumMedium
Key features of public policy since 1980s
LLMLMLLVenezuela
HMMMHHHUruguay
MMMMMMMPeru
LLLLLLMParaguay
LLLLMLMPanama
LMLLMMMNicaragua
HHMMHMHMexico
MMLMMMHHonduras
LMLMLMMGuatemala
HHMMHHHEl Salvador
LLLLLMLEcuador
MMLMMMMDominican Rep
HHHMHMHCosta Rica
HMMMHHHColombia
VHHHHHHHChile
HMMHHHHBrazil
MMMMMHMBolivia
LLMLLMLArgentina
Policy IndexEfficPublic RegCoord / Coherence
Impl/EnfAdaptStabCountry
Key features of public policy since 1980s
LLMLMLLVenezuela
HMMMHHHUruguay
MMMMMMMPeru
LLLLLLMParaguay
LLLLMLMPanama
LMLLMMMNicaragua
HHMMHMHMexico
MMLMMMHHonduras
LMLMLMMGuatemala
HHMMHHHEl Salvador
LLLLLMLEcuador
MMLMMMMDominican Rep
HHHMHMHCosta Rica
HMMMHHHColombia
VHHHHHHHChile
HMMHHHHBrazil
MMMMMHMBolivia
LLMLLMLArgentina
Policy IndexEfficPublic RegCoord / Coherence
Impl/EnfAdaptStabCountry
Key features of public policy since 1980s
LLMLMLLVenezuela
HMMMHHHUruguay
MMMMMMMPeru
LLLLLLMParaguay
LLLLMLMPanama
LMLLMMMNicaragua
HHMMHMHMexico
MMLMMMHHonduras
LMLMLMMGuatemala
HHMMHHHEl Salvador
LLLLLMLEcuador
MMLMMMMDominican Rep
HHHMHMHCosta Rica
HMMMHHHColombia
VHHHHHHHChile
HMMHHHHBrazil
MMMMMHMBolivia
LLMLLMLArgentina
Policy IndexEfficPublic RegCoord / Coherence
Impl/EnfAdaptStabCountry
Why focus on key features of policies?• More naturally linked to the institutional environment than the
content of policies themselves.
• More closely related to political institutions than outcome variables such as economic growth.
• Reassuringly, features of public policy and the overall policy index correlate well with growth and other welfare indicators.
The Policymaking Process (PMP)
• Within our approach, the PMP takes center stage. Characterizing PMP in each country: – Which are the key actors that participate in it?
– What powers and roles do these actors have?
– What preferences, incentives and capabilities do they bring to the table?
– What are the characteristics of the arenas in which they interact?
– How frequent are these interactions?
– What is the nature of the transactions they engage in?
Actors and arenas in policymaking• Background papers study role of various actors and institutions in Latin America’s
PMPs.– Political parties and party systems.– The legislature (and the legislators).– The president and his cabinet.– The bureaucracy and the judiciary.– Sub-national actors.– Business organizations, unions, social movements.– Articulation of technical expertise into the PMP.
• Studies discuss formal and actual roles, interaction with other actors, incentives, policymaking capabilities.
• “Democracies in Development” project also key input for this part of the IPES.
PoliticalInstitutions
PMP Policy
Outcomes
PMP Policy outcomes
• Key insight: features of public policies (such as stability, adaptability, etc) depend crucially on the ability of political actors to strike and enforce inter-temporal deals, and thus achieve cooperative outcomes.
• Cooperation, in turn, depends on factors such as:– Number of actors with substantial impact on policy.
– Temporal horizon (or discount rate) of actors.
– Existence of well functioning arenas for political exchange (eg: institutionalized parties, a well-functioning legislature).
– Credible enforcement technologies (eg, independent judiciary).
Political Institutions PMP• In turn, key aspects of the PMP are determined by the nature of
the political institutions such as constitutional rules, electoral rules, etc, which determine (among others).
– The number of actors and their roles
– Their incentives
– The rules of engagement among them
– The nature of the arenas in which they interact
Different looks at policymaking• IPES looks at links between political institutions, policymaking
processes and policy outcomes from different angles.• Cross-sectional analysis (chapter 6)• Policymaking in action in specific countries (chapter 7)
– Chile: Policymaking in an institutionalized setting– Colombia: the 1991 Constitution and the policymaking process– Brazil and Ecuador: Policymaking in fragmented party systems
• Policymaking in specific sectors (chapters 8 – 11)– Tax policy– Public services– Education– Decentralization, budget processes, and feedback effects
Different looks at policymaking• IPES looks at links between political institutions, policymaking
processes and policy outcomes from different angles.• Cross-sectional analysis (chapter 6)• Policymaking in action in specific countries (chapter 7)
– Chile: Policymaking in an institutionalized setting– Colombia: the 1991 Constitution and the policymaking process– Brazil and Ecuador: Policymaking in fragmented party systems
• Policymaking in specific sectors (chapters 8 – 11)– Tax policy– Public services– Education– Decentralization, budget processes, and feedback effects
Cross-Sectional Analysis
Some results
Some results from cross-section analysis• High-quality policies are associated with:
– Congress with good policy capabilities
– Political parties that are institutionalized and programmatic
– Strong bureaucracies
– Independent judiciaries
• Impact of other institutional variables, such as presidential powers, or degree of proportionality of electoral systems, depends on interactions with other institutional dimensions.
Congressional capabilities• National legislature is natural arena for discussion, negotiation
and enforcement of political agreements.
• Legislators with strong capabilities and legislatures with adequate organizational structures may promote better quality policies (more stable, adaptable, etc).
• We build an index of congressional capabilities, which includes:– Different measures of public perception of Congress
– Education of legislators
– Years of legislative experience
– Specialization in Committees
– A few more subjective variables
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
Congressional Capability Index (1-3 scale)
Po
licy
In
dex
(1-
4 s
cal
e)
Source: Authors' compilation.
Congressional Capabilities and the Quality of Policies
Characteristics of political party systems• In some countries, political parties participate directly in public
policy debate• In others, party systems affect the PMP indirectly, affecting:
– executive-legislative relations– possibilities for coordination in Congress– incentives of elected officials.
• Institutionalized parties , when they are programmatic (competing on basis of public policies and policy outcomes) are more likely to encourage long horizons, and prevent opportunistic behavior.
• They can facilitate inter-temporal bargains within and between parties, since commitments made by current party leaders are more likely to be respected in the future.
Party System Institutionalization, Programmatic Orientation and the Quality of Policies
Venezuela
Uruguay
Peru
Paraguay
Panama
Mexico
Honduras
Guatemala
El Salvador
Ecuador
Dominican Rep.
Costa RicaColombia
Chile
Brazil
Bolivia
Argentina
0
1
2
3
4
5
6
7
8
40 50 60 70 80Institutionalization Index (0-100 scale)
Pro
gra
mm
atic
In
dex
(0-
8 s
cale
)
Policy Index = 2.06
Policy Index = 2.57
Policy Index = 2.04
Note: Boxes show the average value of the Policy Index for the countries in each quadrant.
Source: Authors' compilation and Jones (2005).
Policy Index = "High" Policy Index = "Low" Policy Index = "Medium" Policy Index = "Very high"
Judiciaries and Bureaucracies• Good enforcement and implementation facilitate cooperation,
strengthen quality of policies.
• Judiciary most obvious enforcer. Independent courts ensure that the president does not overstep his boundaries, and that neither branch violates the constitution.
• Inter-temporal cooperation hindered if courts are not independent, actors are not held to their commitments.
• Bureaucracy plays a predominant role in policy implementation.
• But a strong bureaucracy can also facilitate inter-temporal bargains through delegation (eg, central banks, autonomous agencies).
Judicial Independence and the Quality of Policies
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Judicial Independence Index (1-7 Scale)
Po
licy
In
dex
(1-
4 S
cale
)
Source: Authors' compilation and World Economic Forum (2004).
Development of Civil Service and the Quality of Policies
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Development of Civil Service System Index (0-1 scale)
Po
licy
In
dex
(1-
4 s
cale
)
Source: Authors' compilation and Iacoviello and Zuvanic (2005).
Institutional blessings are not independent• “Good” institutions tend to come together. Countries with
stronger congresses tend to have more independent judiciaries.
• Virtuous dynamics: If executives do not tinker with the Courts, this increases their independence and reputation. Independent judiciaries enforce the domain and prerogatives of Congress, increases incentives of legislators to invest in capabilities.
• Vicious dynamics can also occur, unfortunately.
• Problem: strong institutions take time to build, but can be destroyed overnight.
Policymaking in fragmented party systems
Government type and constitutional interruptions
34
15
19
# gov.
Minority
10
5
5
# CI
Majority (or close)
64
28
36
# gov.
4
2
2
# CI
29.429.46.36.3Total
33.37.1 Coalition
26.35.6Single party
%%
Based on 98 democratic governments in Latin America since 1978.CI defined as periods in which either the President or the legislature did not complete their terms.Source: Chasquetti (2004), Payne et al (2002) y and author’s compilation.
Government type and constitutional interruptions
34
15
19
# gov.
Minority
10
5
5
# CI
Majority (or close)
64
28
36
# gov.
4
2
2
# CI
29.46.3Total
33.37.17.1 Coalition
26.35.65.6Single party
%%
Based on 98 democratic governments in Latin America since 1978.CI defined as periods in which either the President or the legislature did not complete their terms.Source: Chasquetti (2004), Payne et al (2002) y and author’s compilation.
Policymaking in fragmented party systems: Brazil and Ecuador
• Political institutions have important common elements
• Most fragmented party systems in Latin America
Source: Jones (2005).
Political Party Fragmentation
1 2 3 4 5 6 7 8
Chile
Honduras
Nicaragua
Dominican Republic
Paraguay
Uruguay
Mexico
Panama
Costa Rica
Argentina
Guatemala
El Salvador
Peru
Venezuela
Colombia
Bolivia
Ecuador
Brazil
Effective Number of Legislative Parties
Brazil and Ecuador: party system fragmentation
ArgentinaHonduras
Chile
Costa Rica
El Salvador
Colombia
Bolivia
Brazil
Ecuador
Uruguay
Venezuela
Paraguay
Panama
Nicaragua
Guatemala
Dominican Republic
Mexico
Peru
10
20
30
40
50
60
0 2 4 6 8 10
Effective Number of Parties
Pre
sid
ent'
s C
ham
ber
Co
nti
ng
ent
Source: Jones (2005), Saiegh (2005).
Policymaking in fragmented party systems: Brazil and Ecuador
• Political institutions have important common elements
• Most fragmented party systems in Latin America
• Strong presidential constitutional powers (to compensate for weak partisan powers)
Source: Payne and others (2002).
Presidential Legislative Powers
0.0 0.2 0.4 0.6 0.8 1.0
Nicaragua
Paraguay
Costa Rica
Bolivia
Mexico
Honduras
Dominican Republic
Guatemala
Venezuela
El Salvador
Uruguay
Panama
Argentina
Peru
Colombia
Ecuador
Brazil
Chile
Presidential Legislative Powers index (0-1 scale)
Policymaking in fragmented party systems: Brazil and Ecuador
• Political institutions have important common elements
• Most fragmented party systems in Latin America
• Strong presidential constitutional powers (to compensate for weak partisan powers)
• Legislators elected under similar electoral rules: open list, PR, incentives to deliver benefits to their jurisdictions.
Policymaking in fragmented party systems: Brazil and Ecuador
• Political institutions have important common elements
• Most fragmented party systems in Latin America
• Strong presidential constitutional powers (to compensate for weak partisan powers)
• Legislators elected under similar electoral rules: open list, PR, incentives to deliver benefits to their jurisdictions.
• Yet completely different political and policy outcomes
Key features of public policy and overall index
LLMLMLLVenezuela
HMMMHHHUruguay
MMMMMMMPeru
LLLLLLMParaguay
LLLLMLMPanama
LMLLMMMNicaragua
HHMMHMHMexico
MMLMMMHHonduras
LMLMLMMGuatemala
HHMMHHHEl Salvador
LLLLLMLEcuador
MMLMMMMDominican Rep
HHHMHMHCosta Rica
HMMMHHHColombia
VHHHHHHHChile
HMMHHHHBrazil
MMMMMHMBolivia
LLMLLMLArgentina
Policy IndexEfficPublic RegCoord / Coherence
Impl/EnfAdaptStabCountry
How can these differences be explained?
• Substantial differences in capabilities of key institutions.
Source: Saiegh (2005)
Lower House Reelection Rates
0 10 20 30 40 50 60 70
Rates of Immediate Reelection to the Lower House (percentage)
Peru (2001)
Argentina (1989-2003)
Guatemala (1994-2004)
Venezuela (1989-2004)
Bolivia (2002)
Ecuador (1996-2002)
El Salvador (2003)
Colombia (1990-98)
Paraguay (1998-2003)
Brazil (1995-2002)
Panama (1999)
Uruguay (1989-99)
Chile (1993-2001)
Judicial Independence Index
1 2 3 4 5
Venezuela
Paraguay
NicaraguaBolivia
Argentina
Ecuador
HondurasPeru
GuatemalaPanama
El Salvador
Colombia
MexicoDominican Rep.
Costa RicaBrazil
Chile
Uruguay
Source: World Economic Forum (2004).
Index of Development of the Civil Service
0.0 0.2 0.4 0.6 0.8 1.0
Panama
El Salvador
Honduras
Paraguay
Peru
Ecuador
Nicaragua
Guatemala
Bolivia
Dominican Rep.
Venezuela
Mexico
Colombia
Uruguay
Costa Rica
Argentina
Chile
Brazil
Source: Iacoviello and Zuvanic (2005).
How can these differences be explained?
• Substantial differences in capabilities of key institutions.
• Presidents in Brazil have necessary tools to engage in political exchanges with other parties, and sustain coalitions in Congress.
How can these differences be explained?
• Substantial differences in capabilities of key institutions.
• Presidents in Brazil have necessary tools to engage in political exchanges with other parties, and sustain coalitions in Congress
• In Ecuador, coalitions are unstable, and tend to collapse as presidential terms progress and elections approach.
Brazil: building coalitions• President needs to rely on coalition to pass his agenda.
• Legislators need to deliver benefits to their communities, add budget amendments involving programs with local benefits.
• Strong presidential powers, include discretional capacity to impound budgetary amendments.
• Key exchange is small investment projects in exchange for political support to pass the president’s agenda.
• One of several inducements president uses to pass agenda in a fragmented system (others include sharing ministries with coalition partners)
Ecuador: ghost coalitions• Legislators have incentives to deliver pork to their communities, but
geographically based items not allowed in the budget.• Being perceived as “gobiernista” is politically costly, so ministries
are not always attractive to potential coalition partners, particularly towards end of presidential term.
• Politicians avoid visible and politically costly commitments, engage in “ghost coalitions”, providing legislative support in narrow set of issues in exchange for immediate benefits.
• Result: unstable coalitions, adversarial relations between the executive and the legislature, and a great deal of political instability.
• This in turn affects time horizons of politicians, and quality of public policies.
Ecuador: Size of Government’s Coalition in Congress over the presidential term (1984-2002)
10
20
30
40
50
60
1 5 9 13 17 21 25 29 33 37 41 45
Administration's months in office
Siz
e o
f th
e P
resi
den
t's
coal
itio
n (
per
cen
t)
Source: Mejía-Acosta (2004).