How Do Political Institutions Work?

Intertemporal Interactions, Veto Players, and Policy Adaptability

Carlos Scartascini

Ernesto Stein

Mariano Tommasi

“ Institutions may be seen as architecture and as rules that determine opportunities and incentives for behavior, inclusion and exclusion of potential players, and structuring the relative ease or difficulty of inducing change, and the mechanisms through which change may be facilitated or denied.”

Rhodes, Binder and Rockman (2006, p. xiii).

Introduction to the Oxford Handbook of Political Institutions

Research on POLITICAL INSTITUTIONSOrigin of institutions

Emphasis on Intertemporal Transactions Williamson: firm North-Weingast: Parliament Fearon: electoral democracy Weingast-Marshall: institutions within Congress

Effects of political institutions (on policies)

Less emphasis on Intertemporal Transactions Mostly (static) spatial models

- Persson and Tabellini: economic dynamics

Some exceptions (institutions within US Congress)- Repeated game versions of Baron-Ferejohn (or of WSJ)

Little of that for 1st order institutions – comparative politics- some informal discussions

- political party systems (Mainwaring)

- Executive / Legislative relations (Shugart, Carey)

Research on POLITICAL INSTITUTIONSOrigin of institutions

Emphasis on Intertemporal Transactions Williamson: firm North-Weingast: Parliament Fearon: electoral democracy Weingast-Marshall: institutions within Congress

Effects of political institutions (on policies)

Less emphasis on Intertemporal Transactions Mostly (static) spatial models

- Persson and Tabellini: economic dynamics

Some exceptions (institutions within US Congress)- Repeated game versions of Baron-Ferejohn (or of WSJ)

Little of that for 1st order institutions – comparative politics- some informal discussions

- political party systems (Mainwaring)

- Executive / Legislative relations (Shugart, Carey)

• GENERAL (AGENDA)

– The effects of institutions on policy should (also) be studied from a perspective of intertemporal transactions

Repeated Games

• SPECIFIC (THIS PAPER)

– Show in one specific and important context how viewing things this way can lead to predictions different from “static” approaches

In particular: VETO PLAYERS Approach

We confront one of their main predictions, that having more veto players necessarily makes policy change less likely.

1. VP Approach

2. IT Approach (& 1 model / example)

3. Empirical Evidence

VETO PLAYERS THEORY

• Tsebelis 1995 1999 2000 2002

– “Tour de force of modern political science theory”– “Veto Players should be required reading for all scholars

of comparative political institutions”– “Veto point and veto player approaches have come to

occupy a central place in comparative politics, especially in the fields of comparative public policy”

– Oxford Handbook of Political Institutions (3 of 2000)– Logic applied to: welfare states, inequality, fiscal

adjustment, tax competition, monetary policy and institutions, international trade arrangements, business environment, European Union policymaking, etc.

VETO PLAYERS THEORYMotivation

• Most previous literature dyadic criteria– presidential / parliamentary– majoritarian / proportional– two party / multiparty– etc

• Comparisons across countries difficult– multiple institutional features

• Interactions? – Cannot just add up the different effects

• Need systemic approach– Tsebelis: “configuration of veto players”

• actors whose agreement is necessary to change policy

VETO PLAYER Approach (Tsebelis 2002)

Dependent variable: • POLICY STABILITY -- Seen as trade-off:

: Gives COMMITMENT : Does not give DECISIVENESS

(capacity to solve problems as they arise)

Independent variable: • (intermediate): VETO PLAYER configuration

– Veto Player: Actors whose agreement is necessary to change policy

• (final) Political Institutions– Tsebelis 2002, Part II: “rules according to which all political institutions

(regime types, parliaments, party systems, parties, and so on) are translated into a series of veto players.”

Prediction: More veto players Policy more stable

NOTE

Most of Tsebelis’ initial work was mainly on Parliamentary systems

POLITICAL INSTITUTIONS

VETO PLAYER CONFIGURATION

POLICY STABILITY

Size of winset

Presidents, Parliaments, and Policy

Haggard and McCubbins (eds.) & Ch 1 / Cox and McCubbins Ch 2 /Shugart and Haggard Ch 3 / countryXpolicy case studies

• Instruments a “veto player logic” to presidential democracies• Refines specification of dependent variable

Political Institutions

Constitutional Division of Power Presidentialism Bicameralism Federalism Judicial Review

Electoral System Effects on Party System (fragmentation)

Veto Player Configuration

- Separation of Power

- Separation of Purpose

Effective Number of Veto Players

Characteristics of Policymaking/of Policies

• Decisiveness:Ability to change policy

• Resoluteness:Ability to commit to a given policy

Presidents, Parliaments, and Policy

Haggard and McCubbins (eds.) & Ch 1 / Cox and McCubbins Ch 2 /Shugart and Haggard Ch 3 / countryXpolicy case studies

• Instruments a “veto player logic” to presidential democracies• Refines specification of dependent variable

Political Institutions

Constitutional Division of Power Presidentialism Bicameralism Federalism Judicial Review

Electoral System Effects on Party System (fragmentation)

Veto Player Configuration

- Separation of Power

- Separation of Purpose

Effective Number of Veto Players

Characteristics of Policymaking/of Policies

• Decisiveness:Ability to change policy

• Resoluteness:Ability to commit to a given policy

Haggard and McCubbins, p. 6:A more decisive polity must necessarily be less resolute

Summary “Veto Player Theories”

PROPOSITION 1:

A more decisive polity must necessarily be less resolute (Haggard and Mc Cubbins, p. 6)

PROPOSITION 2:

As the effective number of vetoes increases, the polity becomes more resolute and less decisive

(Haggard and Mc Cubbins, p. 27)

Many veto players make significant policy changes difficult or impossible (Tsebelis 2002, p. XV).

Summary “Veto Player Theories”

PROPOSITION 1:

A more decisive polity must necessarily be less resolute (Haggard and Mc Cubbins, p. 6)

PROPOSITION 2:

As the effective number of vetoes increases, the polity becomes more resolute and less decisive

(Haggard and Mc Cubbins, p. 27)

Many veto players make significant policy changes difficult or impossible (Tsebelis 2002, p. XV).

In this paper we debate these propositions.

AN INTERTEMPORAL FRAMEWORK

POLITICAL INSTITUTIONS POLICY STABILITY

VETO PLAYER CONFIGURATION

NumberIdeological DistanceInternal Cohesion

POLITICAL INSTITUTIONS

“VETO PLAYER” CONFIGURATION

Separation of PowerSeparation of Purpose

DECISIVENESS

RESOLUTENESS

POLITICAL INSTITUTIONS

Intertemporal cooperation

structure

ADAPTABILITY

STABILITY

A. VP Framework of Tsebelis (1995 and 2002)Independent variables Theoretical construct Dependent

variable/s

B. Framework of Cox and McCubbins in Haggard and McCubbins (2001)Independent variables Theoretical construct Dependent

variable/s

C. Intertemporal cooperation Framework Independent variables Theoretical construct Dependent variable/s

Decisiveness

Resoluteness

Decisiveness

Resoluteness

Decisiveness

Resoluteness

Less cooperative policymaking environment

More cooperative policymaking environment

Framework: Intertemporal Political Cooperation

• Framework emphasizes ability of political actors to cooperate

• Policies with some desirable characteristics likely to emerge if

players can – cooperate

– uphold agreements over time

• In systems that encourage cooperation– consensus on policy orientation more likely to emerge

• related to notions of consensual democracy– related to notions of deliberative democracy

– successive administrations more likely to build upon achievements of predecessors

– easier to adjust to changing circumstances

• Whether political systems encourage/achieve cooperation– Depends on various things

– Some important ones relate to political institutions

• In previous work we have – Related some aspects of political institutions to better policies

• Political parties that are institutionalized and programmatic

• Legislatures with sound policymaking capabilities

• Independent judiciaries

• Strong bureaucracies

– Studied the effect of some specific institutions usually considered

• This paper focus on (“reduced form”) number of veto players– Sometimes it might facilitate cooperation and hence, policy adjustment

Modelling Intertemporal Political Cooperation

• Nothing new• Similar modelling as Dixit 2003, Dixit et al 2000, Alesina 1988, de Figueiredo 2002• PM as intertemporal exchanges (TCETCP) as in North 1990, Dixit 1996, Levy and Spiller

1996, Huber and Shipan 2002, Epstein and O’Halloran 1999.

FAMILY OF MODELS:

1. Politics and policymaking take place over time (repeated game)

2. Shifting power over time (t)

3. Conflict and commonality of interests (non-trivial payoff matrix)

4. Changing underlying conditions (t)

Dependent variable: policy characteristics Y, including:– Adaptability: Y*()

– Stability: Y does not change for reasons other than (invariant to )

Defining the dependent variablesTsebelis: Stability. Cox McCubbins: Decisiveness – Resoluteness.

This paper:Adaptability:

• the ability to adapt to exogenous shocks (Tsebelis 1999: 591)

• the flexibility to resolve problems faster (Tsebelis 2002: 7)

• the capacity to solve problems when they arise (Tsebelis 2002: 6-7)

• the ability to innovate when policies fail (Weaver and Rockman 1993)

• the adaptation of social programs to contemporary socio-economic risks (Natali and Rhodes, 2006)

• the ability of policy to adapt to new conditions (International Institute for Sustainable Development, 2007)

Stability:• the ability to ensure policy stability so that policies have time to work (Weaver and Rockman

1993: 6)

• the ability to make and maintain international commitments in the realm of trade and national defense (Weaver and Rockman 1993: 6)

• credible commitment not to alter the rules of the game each time there is a government change

• the ability to uphold promises (Cox and McCubbins 2001: 31)

• the ability to carry out a policy compromise (Cox and McCubbins 2001: 31)

• the consistency with which a policy is implemented (Rodrik 1995: x)

The model• We write one specific model within the general “class” above

• Example could be interpreted as ( )– budget

• negative shock that requires risk sharing• positive investment opportunity / externalities

– (non-budget) independent random issues arising over time, possibility of changing from a given status quo

Repeated game• δ• N players, v of which seat at decision table t : assigns who are at the v players at the table at t

• Unanimity (v veto players)• Allocate budget

• t

Non-cooperative equilibrium• Ignore welfare of (N-v) outside the table at that period

– Exactly how split within table depends on bargaining protocolY()

First best • Max everybody’s (N) long term welfare

– responsive to t

Y*()

Cooperative equil that implements first best• Trigger strategy: Do as first-best unless an allocation different from

Y*() was implemented. If so revert to non-coop foreverY*()

Non-cooperative equilibrium• Ignore welfare of (N-v) outside the table at that period

– Exactly how split within table depends on bargaining protocolY()

neither adaptable nor stable

First best • Max everybody’s (N) long term welfare

– responsive to t

Y*()

Cooperative equil that implements first best• Trigger strategy: Do as first-best unless an allocation different from Y*()

was implemented. If so revert to non-coop foreverY*()

adaptable and stable

• Verifying under what conditions deviating from this equilibrium is not profitable:

• Cooperation is more likely for larger v

• In this example, cooperation, and hence stability and adaptability, increase with the number of veto players.

vNN 11

11

Other channels

• General point: the optic of intertemporal cooperation generates additional channels for the effects of political institutions on outcomes of interest

• This paper focuses on v

• We built one specific example in which greater v increases cooperation (and hence stability and adaptability)

• There are other channels through which that might happen:

– What matters is not only

Institutions how many people at the decision table today,

but also

Institutions intertemporal allocation of decision power

(what are my chances of being there tomorrow)

And that could affect my willigness to cooperate today, allowing a change that

might harm me in the short run

Veto Player TheoriesPROPOSITION 1:

A more decisive polity must necessarily be less resolute

PROPOSITION 2:

As the effective number of vetoes increases, the polity becomes more resolute and less decisive

Intertemporal ApproachPROPOSITION 1’

A more decisive polity not necessarily must be less resolute. There are some forces (of different equilibria in repeated-interaction contexts) leading to a positive association between decisiveness and resoluteness (adaptability and stability).

PROPOSITION 2’ Many veto players do not necessarily make significant policy changes difficult or impossible. There are some channels through which more veto players increase policy adaptability.

6. Empirical Analysis

6.1. Dependent Variables

• Indicators of Stability and Adaptability

• The best we could get our hands on from wide international data sets

• Mostly subjetive data: opinions of businessmen, experts

• Various different sources

• We are exploring the correlation of these indexes with more specific (sometimes objective) measures of policy characteristics (available for smaller samples)

• Cross-section

Box 1 :Correlation Adaptability Index and degree to which fiscal and social policy respond to economic and social situation.

SocialSpending

TotalSpending

Social Spending 1.00Total Spending 0.6961* 1.00Adaptability 0.5289* 0.4894*Adaptability-hi 0.5928* 0.5324*

(source: Braun and di Gresia 2002)

Dependent VariablesAdaptability: Four variables:• Bertelsmann Transformation Index (BTI) Adaptability Variable 2006 ranking. Questions comprising BTI’s

adaptability index include the ability of the political leadership to act flexibly, political leaders’ capability for learning, and whether political leaders can replace failing measures with innovative policy.

• Columbia State Capacity Survey, (1990, 1999, 2000, 2002) Rate the state’s ability to respond effectively to domestic economic problems.

• Profils Institutionnels-database 2006: Decision-making capacity of the political authorities in economic matters (responsibility, rapidity, etc)

• Columbia University State Capacity Survey (1990, 1999, 2000, 2002) Rate the state’s ability to formulate and implement national policy initiatives

Stability: Four variables:• Standard deviation of the detrended Fraser Index of Economic Freedom (quadratic trend) (1980-2004)• Global Competitiveness Report (GCR, 2002): Legal or political changes over the past five years have

(1=severely undermined your firm's planning capacity, 7=had no effect)• Global Competitiveness Report (GCR, 1998, 1999, 2000, 2002): New governments honor the contractual

commitments and obligations of previous regimes (1=not true, 7=true). • Profils Institutionnels-database 2006: “Consistency and continuity of government action in economic matters”

6.2. Testing Veto Players Hypotheses

PROPOSITION 1: A more decisive polity must necessarily be less resolute

Contrary to the hypothesis, all 17 correlations are positive, many of them significant.(controlling for level of development, region and legal origin)

Table 2. Correlation between the components of stability and adaptability –no controlsAdaptability Components

StabilityComponents

Adaptability(BTI)

StateResponsiveness

DecisionMakingCapacity

StateEffectiveness Adaptability

0.1883

0.41***93

0.31***73

0.42***93

Frazer Volatility0.0870

0.32***83

0.23*65

0.29***83

0.1049

0.72***61

0.41***54

0.61***61Government

Commitment 0.2239

0.37***53

0.1546

0.29**53

0.35***49

0.62***61

0.62***53

0.59***61

Policy Changes0.40**

390.2252

0.47***45

0.26*52

0.35***65

0.47***78

0.74***85

0.51***78Policy

Consistency 0.41***49

0.1363

0.50***69

0.28**63

0.54***122

Stability0.16100

Note: significant at 1% level (***), 5% (**), and 10% (*)Second line in each row is number of observationsSecond row for each variable is the correlation controlling for Ln(GDPpc), Region, and Legal Origin

Stability and Adaptability Scatterplot

-2-1

01

Sta

bili

ty/G

DP

, Reg

, an

d Le

gal

-2 -1 0 1 2Adaptability/GDP, Reg, and Legal

PROPOSITION 2:

“… more veto players reduce policy adaptability”

We utilized four different measures/proxies for the number of veto players that were available in broad cross national empirical studies:

Table 3. The veto variablesVariable Coverage Observations Source

Executive Constraints (Polity IV)It refers to the extent of institutionalizedconstraints on the decision-making powers ofchief executives, whether individuals orcollectivities.

average of datafrom 1990-

2003147

University ofMaryland Polity IVProject

Executive Constraints (Henisz):It refers to the limit on the executive's actions.(1) Unlimited Authority: There are no regularlimitations on the executive's actions, to (3)Slight to Moderate Limitations an ExecutiveAuthority, to (7) Executive Parity orSubordination: Accountability groups haveeffective authority equal to or greater than theexecutive inmost areas of activity. Thosepolities described with a (3) or above alsocontain an independent judiciary

average of datafrom 1990-

2004147

Political ConstraintsIndex Dataset

POLCONV:It is constructed by identifying the number ofindependent branches of government with vetopower over policy change and is then modifiedto take into account the extent of alignmentacross branches of government using data onthe party composition of the executive andlegislative branches.

average of datafrom 1990-

2004151

Political ConstraintsIndex Dataset

Checks:This variable adds the number of checks acountry has. Starting with one in countrieswhere legislatures are not competitivelyelected, it increments its value if there is a chiefexecutive, if the chief executive iscompetitively elected, if the opposition controlsthe legislature, and so on.

average of datafrom 1990-

2004148

Database of PoliticalInstitutions,

Table 4. Regressions on adaptability (weighted least squares).(1) (2) (3) (4) (5) (6) (7) (8)

ExecutiveConstraints(Polity IV)

0.14***(0.04)

0.16***(0.05)

ExecutiveConstraints(Henisz)

0.14***(0.04)

0.17***(0.05)

PolconV 1.07***(0.29)

0.94***(0.32)

Checks 0.12**(0.05)

0.08(0.06)

Ln(GDPpc) Yes Yes Yes Yes Yes Yes Yes YesRegion Yes Yes Yes YesLegal Origin Yes Yes Yes YesAdjR2 0.43 0.43 0.44 0.40 0.63 0.61 0.61 0.45Obs 107 108 109 109 106 107 108 106

Note: significant at 1% level (***), 5% (**), and 10% (*)

6.3. Some evidence incorporating intertemporal factors

Table 5. Intertemporal cooperation, and stability and adaptability (weighted least squares). Stability Adaptability (1) (2) (3) (4) (5) (6) (7) (8) Time Horizon 0.09**

(0.04) 0.22***

(0.07)

Cohesion 0.01 (0.04)

0.20*** (0.07)

Party Institut. 0.20*** (0.07)

0.42*** (0.12)

Judicial Independence

0.31*** (0.06)

0.54*** (0.09)

Ln(GDPpc) Yes Yes Yes Yes Yes Yes Yes Yes

Region Yes Yes Yes Yes Yes Yes Yes Yes

Legal Origin Yes Yes Yes Yes Yes Yes Yes Yes

AdjR2 .53 .52 0.55 0.64 .50 .61 0.63 0.62

Obs 103 75 104 100 107 73 108 104

Note: significant at 1% level (***), 5% (**), and 10% (*)

6.4. Attempting to recover the veto-players trade off

Decisiveness

Resoluteness

Less cooperative policymaking environment

More cooperative policymaking environment

Hypothesis 1

Table 6. Correlation between adaptability and stabilitySample divided by

Time horizon1 Time horizon2

Controlling forLn(GDPpc) + Regional +Legal Origin

Whole sample High Low High Low

0.16 0.62*** -0.22 0.45*** -0.61***

N 100 50 50 78 21Note: significant at 1% level (***), 5% (**), and 10% (*)1 Low and high corresponds to countries being above or below the median2 Low and high corresponds to countries being above or below the 75th percentile

Hypothesis 2

Table 7. Regressions on adaptability (weighted least squares). (1) (2) (3) (4) Executive Constraints (Polity IV) 0.08

(0.08)

Executive Constraints (Henisz) .09 (.08)

PolconV .18 (.42)

Checks .01 (.06)

Time horizon .09 (0.09)

.09 (.09)

.09 (.09)

.08 (.09)

Cohesion .18*** (.06)

.18*** (.06)

.18*** (.06)

.18** (.07)

Party Institutionalization .25* (.14)

.26* (.14)

.24* (.14)

.23 (.14)

Judicial Independence .32** (.13)

.31** (.13)

.36*** (.13)

.40*** (.11)

Ln(GDPpc) Yes Yes Yes Yes Region Yes Yes Yes Yes Legal Origin Yes Yes Yes Yes AdjR2 0.72 0.72 0.71 0.71 Obs 69 69 69 68

Note: significant at 1% level (***), 5% (**), and 10% (*)

Parting thoughts

• Intertemporal cooperation is important

• Keep that in mind if you have to design institutions