Bringing the Legislature Back In: Examining the Structural Effects of National Legislatures on Effective Democratic Governance

Abstract

Extant research in comparative politics has examined the role of institutional frameworks such as constitutional design, the nature of the electoral systems, parliamentarism and federalism on the quality of governance. Understanding variations on effective democratic governance has assumed a state-centric approach that has largely neglected how strong legislatures can drastically affect political outcomes. This study finds empirical evidence that the strength of national legislatures (in terms of its influence over the executive, institutional autonomy, its specified powers and institutional capacity) is correlated to effective democratic governance as measured by voice and accountability, governmental effectiveness, regulatory quality, control of corruption and rule of law entrenchment based on a cross-national analysis of 150 countries with available data from the period 1996–2016. The results hold even when the sample is restricted to developing countries, where party systems are more likely to be under-institutionalized. A sensitivity analysis also confirms that the relationship between strong legislatures and effective democratic governance is not attenuated or conditioned by its interactive effect with other institutional arrangements. Implications suggest that the substantive strength of national legislatures promotes higher levels of democratic accountability, and that the international community must focus on frameworks that strengthen global legislatures to avert political instability and creeping authoritarianism.

Empirical studies that investigate the determinants behind good governance have examined various institutional dynamics that promote it (Andrews and Montinola 2004; Congleton and Swedenborg 2006; Doorenspleet 2005; Faguet 2014; Gerring et al. 2009). A significant omission in these extant studies is understanding how the substantive strength of national legislatures can have a long-standing positive effect on effective governance through three mechanisms: promoting vertical accountability and representation, responsive and efficient legislation, and providing constraints on the executive branch that promotes horizontal accountability (Olson 1994). The state-centric approach of examining governance has largely neglected the influence of legislative power and how strong legislatures not only promote checks and balances but also have long-term effects on how governments function effectively. This critical omission is remarkable considering how policymakers, foreign policy experts and international organizations such as the United Nations Development Program (UNDP) and World Bank have explored and analysed how the strength of legislatures in both the developed and the developing world positively impact governance.

More importantly, practitioners have itemized the role of expanding oversight powers, public account committees and the legislature's autonomous nature in curtailing governmental inefficiency, curbing corruption and enhancing democratic representation (McGee 2002; UNDP 2011; United Nations Economic Commission 2011; World Bank Institute 2013; Yamamoto 2007). Policy analysts have therefore prioritized legislative-strengthening programmes as important in promoting macro-economic stability among emerging democracies and are crucial in averting democratic reversals (Arnold 2012; Barkan 2008; Center for Democracy and Governance 2000; Guinn and Straussman 2017). According to Goldstone and Ulfelder (2004: 19), stable democracies that endure arise because they develop institutions that impose clear limits on executive authority. This sole prerogative rests within and is the responsibility of a strong and independent legislature. Case-study evidence in Africa illustrates how strong legislatures promote the institutionalization of party systems that lead to effective governance (Barkan 2009: 2; Salih 2007: 14), while in Latin America some national legislatures historically have engaged in ‘originative, proactive, and reactive’ attempts to challenge the executive to ensure efficacious policymaking (Cox and Morgenstern 2001: 171; Morgenstern and Nacif 2002). Hence, the objective of this study is to investigate a conundrum that practitioners have advocated for, but scholars have neglected – that is, systematically and empirically to assess cross-nationally this significant question: do strong legislatures promote effective democratic governance?

There is an underlying gap in understanding the nexus between strong legislative institutions and effective democratic governance. Although seminal works in the past by Stapenhurst and Pelizzo (2002) and Pelizzo and Stapenhurst (2004, 2013, 2014) lay out the theoretical, empirical and normative foundations for the positive link between the oversight responsibilities of legislatures and the quality of democracy with several case studies in the developing world, there is still a systematic lack of empirical attention to investigating the relationship of how the substantive strength of legislatures affects all aspects of democratic governance. To wit, the lack of scholarly attention placed on legislatures and their purported effect on governance is attributed to a lack of cross-national measures that approximate the relative strength of national assemblies that is comparable across nation states (Munck 2009: 140).

This oversight is based on the dominant paradigm that legislatures are constrained from performing their constitutional duties because of institutional design. It is assumed that presidential systems have stronger legislatures in consolidated democracies because of the institutionalization of checks and balances, while in nascent presidential democracies legislatures are weak because of their penchant for electing populist leaders that use their charismatic appeal and weak constitutions to thwart the powers of other branches of government (Elgie 2012; O'Donnell 1994). Likewise, parliamentary systems are known to promote higher levels of political accountability because strong majoritarian governments allow voters to make better-informed choices in evaluating the performance of incumbents because of clarity of responsibility and stronger levels of party organization (Gerring and Thacker 2004). A flaw in this paradigm is that nation states do not entirely fit the classification of having a pure presidential or parliamentary system. As Fish and Kroenig (2009: 5) argue, ‘Such pure types are rare. Outside of Europe and North America, it is difficult to find pure parliamentary, semi-parliamentary, or presidential systems.’ Alan Siaroff (2003) posits that simplistic distinctions delineating presidential systems from semi-presidential and parliamentary systems prove problematic conceptually because prime ministers are usually accountable to different branches of government and have veto powers and oversight responsibilities that equate or even rival presidential systems. Thus, scholars need to refine the conceptual definitions of forms of governmental structure or focus more on other institutional arrangements that matter, such as the intrinsic strength of the legislature that can have long-term effects on governance.

Understanding how legislatures affect policy outcomes has been crowded out by scholarship demonstrating how the evolution and rise of political parties and interest groups have made the legislature an insignificant body of lawmaking (Cox and McCubbins 2005); how the rise of welfare states and increasing bureaucratization have made the executive branch more influential as a body that can affect all aspects of policymaking (Richardson and Jordan 1985); how legislatures in the developing world lack professionalization and are marginal in policymaking overpowered by strong executive structures (Mezey 1983: 512); how public opinion over time has rendered the legislature as a body that is incapable of policy change, reflected in public opinion polls, low trust ratings among the public and dismal voter turn-out rates in legislative elections (Loewenberg 2015); and how globalization and the rise of supranational institutions have rendered legislatures ineffective in influencing domestic politics (Sassen 1996). Thus, comparative research on the transformative power of legislatures is lacking, undervalued and stale since ‘they are poorly understood and appreciated by democratic publics across the world’ (Martin et al. 2014: 4), which played a role in how academics increasingly discounted the legislature's political relevance. Relegating legislative strength in understanding effective governance is surprising considering that there is substantial empirical evidence that legislatures are not irrelevant political actors, at least in the context of how coalition governments within legislatures exert influence on domestic political outcomes (Carroll and Cox 2012; Kim and Loewenberg 2005; König 2006). Martin and Vanberg (2011: 5) clearly demonstrate that European countries that have strong legislative committees allow for coalition parties to shape policy decisions that are of national importance. Thus, executive structures in policymaking do not necessarily predominate; legislatures are not mere rubber-stamping appendages but have agency.

This study posits that the strength of the national legislatures is more relevant than other institutional variables in understanding why some states provide better governance than others. Legislatures are, after all, considered to be the ‘bedrock of democratic governments’ and the paramount institution ‘that is responsible for lawmaking upon which society is governed’ (Olson 1994: 1–2). After all, it is national legislatures that enact laws that affect the efficacious provision of public goods, the quality and implementation of laws that concern the national interest, the entrenchment of policies that ensure macro-economic stability, and the passage of national directives that institutionalize the rule of law, anti-corruption programmes and protocols that protect citizens from civil rights violations and guarantee due process (Loewenberg et al. 2002). Even among hybrid and authoritarian regimes, legislatures often challenge state authority on governmental malfeasance through oversight mechanisms, especially when members of the opposition want to establish political reform (Barkan 2008).

It is important to note that legislative strength can be endogenous with good governance in itself. Effective governance can generate causal pathways that promote efficacious institutions. However, extant literature from prominent scholars provides evidence that legislatures precede effective governance in affecting political outcomes. First, O'Donnell (1999: 77–78) posits that national legislatures are the primary institutions that promote ‘horizontal’ accountability that induces good governance – more so than other institutional variables like electoral system choice, the form and mode of government (pertaining to decentralization or adopting presidentialism versus parliamentarianism), party system strength and the extent of state capacity (Messick 2002).

More significantly, the degree of legislative strength is consequential in promoting effective democratic governance across time because many democracies suffer from judicial quiescence and underdeveloped party systems (Fish 2006: 12–13). As such, legislatures are pre-eminent institutions that affect efficacious governance even in polities with weak state capacity. Legislatures are also identified as incubators that build and sustain strong party systems, efficient bureaucracies and civic associations that can enhance higher levels of political accountability (Olson 1994). Further, legislative strength temporally precedes other institutional variables of significance in influencing variations on policy outcomes as they historically determine through bargaining how much political power is allocated to the executive over time, the extent to which other branches of government are constrained in terms of raw power and whether the legislature exhibits ‘means’ independence that conditions the extent and scope of judicial review by the courts (Olson and Norton 1996; Thompson and Silbey 1984). This is particularly evident in both the developing world and Western industrialized democracies (Mezey 1978, 1983). Lastly, legislative strengthening in terms of how the legislature develops substantive power is a process that historically accrues gradually over time (Guinn and Straussman 2017). Thus, we can temporally identify nations that have strong legislative institutions versus those that have weaker ones across time, which can have long-standing direct effects on effective governance and modern statecraft building (Arter 2013). This even applies to authoritarian polities where binding legislatures are seen to have a positive effect on policy outcomes such as economic growth and domestic investment (Wright 2008).

The study proceeds as follows. The first part discusses how existing studies illustrating how national legislatures affect the quality of government cross-nationally are scant, necessitating a systemic, empirical cross-national study specifying how powerful legislatures affect political accountability. The second part provides a theoretical overview of how nation states with strong legislatures are likely to have higher levels of: (1) democratic representation; (2) policy outputs (through active legislation); and (3) oversight. All these elements promote efficacious governance. The study then proceeds to operationalize the dependent and independent variables with a discussion of how measuring legislative strength has conceptual deficiencies and limitations that need to be addressed. The next part explains the research design, followed by the presentation of empirical results and a discussion of the implications of the findings.

Legislatures and effective governance

There is scant empirical work that examines the effect of legislatures on the effectiveness of governance. One of the few studies that explores how legislatures affect democratic governance is Fish's (2006) seminal analysis of 25 post-communist polities that transitioned during the 1990s. He finds that the powers accorded to the respective national assemblies (as stipulated in each country's founding democratic constitutions) are correlated to higher levels of basic freedoms and civil liberties (based on the Freedom House index). Legislative strength among post-communist polities facilitated the institutionalization of party systems that ensured democratic consolidation. The evidence points to how legislative weakness may inhibit democratization ‘by undermining the development of political parties’ (Fish 2006: 12).

Strong legislatures facilitate lower levels of corruption – a dimension of effective governance. Using a global sample of all countries with a population of greater than 500,000 from 2000 to 2010, Fish et al. (2015) found that the ability of the legislatures to constrain the executive, exercise operational autonomy and rival the executive in making laws is associated with lower levels of executive-level corruption. Their work was based on expert-level surveys concerning executive-level bribery and instances of embezzlement and theft. Although Fish et al.'s study is cross-national in scope, the empirical analysis only controls for GDP per capita, oil exports per capita and British colonial legacy, which weakens the generalizability of the findings because of the possibility of omitted-variable bias.

The extent to which legislatures are specialized appears to have a positive effect on the efficiency of the bureaucracy. It is found that the degree of legislative specialization has a positive effect on bureaucratic effectiveness in a global sample of 74 countries, and the effect is more pronounced in countries that employ a presidential system (Jeong 2016: 1104). Legislatures that have multiple legislative committees tasked with specialized tasks are better suited to ‘use ex ante statutory control and ex post oversight of bureaucracies’ (Jeong 2016: 1095). This is because legislatures with powerful committees are more likely to generate opposition parties that will exercise their oversight responsibilities as concrete policy issues are addressed and debated, compared to legislatures where such committees are amorphous – which makes it likely that legislatures will not challenge the policies of incumbents (Jeong 2016).

Increasing the breadth and scope of oversight mechanisms within legislatures can decrease opportunistic behaviour among politicians and improve governance (Kaufmann and Dininio 2006). Using a global survey employed in 82 national legislatures, Stapenhurst et al. (2014) operationalize legislative oversight based on parliamentary tools that can scrutinize the behaviour of governmental activities, including the presence of ‘ombudsmen, auditing institutions, specialized parliamentary committees, public hearings, and interpellations that may end with a vote in the chamber’ (Stapenhurst et al. 2014: 290). Based on their empirical analysis, legislatures that have more oversight functions are correlated to lower levels of perceived state corruption based on Transparency International's Corruption Perception Index using 2012 data, although the relationship is marginally significant at (p < 0.10) and does not control for levels of economic development and regime type.

Finally, legislative strength operationalized as rules imposed on its members can have a drastic effect on levels of corruption. From a sample of 64 developing countries between 1984 and 2004, it is found that the presence of rigid agenda setting by dominant parties and the forced expulsion of dissidents in legislatures are likely to increase corruption perceptions (which are statistically significant at p < 0.01 across all model specifications), using data provided by Transparency International and the International Country Risk Guide group (Yadav 2012: 1047).

Theoretical considerations and hypotheses

Previous studies have linked the improving quality of legislatures to improving levels of democracy, enhancing bureaucratic effectiveness and affecting the prevalence of state-level corruption. A key limitation of these studies is that they restrict the analysis to a cross-national study within a single time frame, or they limit the cases to post-communist countries, developing states or countries with available data – constraining the analysis to a small-N of less than 100 countries. Hence, we cannot discern if these studies have implicit generalizability. Apart from Vineeta Yadav's (2012) and Fish et al.'s analyses (2015), which focus on how legislatures promote and inhibit corruption, extant work lacks explanatory leverage in that it fails to examine the perceived relationship between legislative quality and effective governance across time. Furthermore, the way legislative measures were operationalized in previous studies is restrictive regarding what specific legislative powers are present or absent within a given polity. They fail to account for the dynamic or holistic quality of legislative strength based on multiple dimensions centred on a legislature's main responsibilities of representation, effective legislation and oversight (Loewenberg 2015). By parsing out the quality of legislatures based on each of these responsibilities, previous studies fail to consider that these facets of legislative responsibilities are all interrelated, such that if one legislative chamber does not truly represent citizen interests, it may affect its ability to formulate effective public policies and thus will diminish its capacity to perform checks or constraints on the executive.

This study argues that the quality of the legislature should employ the total quantified powers it has, based on multiple dimensions. Strong legislatures are conceptualized as those that have influence over the executive branch, those which have institutional autonomy, those whose powers are clearly specified in their respective constitutions and those that have the institutional capacity to perform their tasks (Fish and Kroenig 2009: 4). As an aggregate, nation states that have legislatures that score highly on all these dimensions should have direct effects on promoting effective democratic governance by: (1) improving or enhancing political accountability within the political system and empowering citizen's rights (voice and accountability); (2) producing effective public services, an efficient civil service, strong contract enforcement and a government committed in respecting private transactions among citizens and protecting private property (governmental effectiveness); and (3) preventing and controlling governmental malfeasance where public officials are less likely to circumvent the rule of law (control of corruption and the entrenchment of the rule of law).¹

Legislative strength and voice

Stronger legislatures lead to higher levels of voice accountability through the intrinsic or self-interested motivation of legislators to be re-elected (Collord 2016; Mayhew 1974; Opalo 2019), their accessibility to the public (Johnson 2005), their ability to aggregate and articulate citizen demands through ‘means’ autonomy, and converting them to actionable legislation (Barkan 2008) and building citizen–legislator linkages that make them responsive to public opinion and citizen demands (Hibbing 1988; Polsby 1968). As such, the legislature's influence over the executive, its specified powers, degree of institutional autonomy and institutional quality can have a direct bearing on positively affecting effective democratic governance (Guinn and Straussman 2017). In short, stronger legislatures should lead to higher levels of citizen empowerment (Fish 2006).

International organizations and country donors have worked to enhance the institutional capacity of legislatures by strengthening the legislature's ability to enhance its representation function through several outreach programmes. These programmes range from information access through media advertisements, increasing ombudsperson work for legislators to enhance linkages and relationships with citizens, and formalizing opportunities for citizens to meet, question and petition their district representatives through popular initiative programmes (Guinn and Straussman 2017). USAID encouraged public participation in the legislative process by increasing Citizen–Parliament Connections (CPCs), where strong legislatures are construed as those that can ‘respond to the demands of the citizenry, identify, and tackle social problems through legislation, and monitor the government bureaucracy in the public interest’ (Arnold 2012: 441).

Democracies that adopted legislative institutionalization programmes also allocated discretionary funds to increase constituency service in the rural regions. This process allowed for higher levels of citizen representation as it ‘achieved a balance between the internal workings of the lower house and its commitment to community members’ (Arnold 2012: 453). Further, some legislatures in Latin America became more professionalized than others and were able to allocate discretionary funding to members of parliament to hire or contract policy experts or personal secretaries, which increased contact hours between legislators and constituents. This strengthened the quality of representation in Chile, Mexico and Brazil, allowing such countries to attain higher levels of political accountability. This compares with other Latin American polities that leave it up to the personal budgets of MPs to hire staff, drastically reducing legislative outreach to citizens and negating citizenship empowerment (Arnold 2012). As such, weaker legislatures in Paraguay, Bolivia and Ecuador, which are not as institutionalized as others, are beset with political accountability issues and struggle with promoting vertical accountability and transparency (Alemán and Tsebelis 2016; Doyle 2020).

This leads to the formulation of the first hypothesis:

Hypothesis 1: Strong legislatures are correlated with higher levels of citizenship empowerment and voice and accountability.

Legislative strength and governmental effectiveness

The strength of legislatures, which is contingently based on their institutional independence, the specified powers they possess granted by the constitution and their institutional quality, can have a long-standing effect on governmental effectiveness (Saiegh 2005). The causal mechanisms at play are such that in some countries the legislature is the main source of policymaking (Olson 1994: 6–7) and the main purveyor of statutory and administrative law. In these cases this can involve the efficient provision of public services, the functioning and organization of the civil service and the allocation of public finances for infrastructure development deemed important by the general population, as part of their national directive (Blomgren and Rozenberg 2015). In some polities, the legislature is also the chief entity that implements and establishes regulatory policy concerning the distribution and provision of public goods (Ashworth and de Mesquita 2006: n.1; Battaglini et al. 2012: 407–408; Fréchette et al. 2012). In other countries, these responsibilities are not the prerogative of the legislature and are directly granted to the executive branch, whose unrestrained powers can divert resources away from public services, thus curtailing effective governance. Furthermore, stronger national assemblies that have ‘means’ autonomy and more specified powers allow for increased partisan competition among legislators that tend to generate higher levels of fiscal transparency, which contributes to the efficient allocation of the national budget for public goods provision (Wehner and De Renzio 2013).

Stronger legislatures with real autonomy are considered vital in combating the inefficient distribution of public goods and services (Stapenhurst and Pelizzo 2002). Barkan (2009) advances the argument that even in nation states that have weak state capacity, as in Africa, strong legislatures are on the rise in terms of affecting the trajectory of regulatory and economic policies that facilitate the formation of effective bureaucracies, as can be seen in Kenya, while weak legislatures drastically weaken the institutional environment to enhance the quality of public services, as evident in Mali (van Vliet 2014). Likewise, several international organizations have targeted increasing financial assistance to legislative-strengthening programmes in the developing world because of case-study evidence suggesting that strong legislatures have a direct role in sustaining sound policies that promote state capacity, sustainable economic growth, efficient bureaucracies, contract enforcement, protection of property rights and larger domestic investments in human capital (World Bank Institute 2013).

This leads to the formulation of Hypotheses 2a and 2b:

Hypothesis 2a: Strong legislatures are correlated with higher levels of governmental effectiveness.

Hypothesis 2b: Strong legislatures are correlated with higher levels of regulatory quality reflecting sound policies that implement and enforce compliance to contracts and the respect of economic and property rights, as well as private sector development (construed as a public good).

Legislative strength and controlling governmental malfeasance

Overall, it is expected that strong legislatures that exhibit substantive strength based on their influence over the executive, institutional autonomy, specified powers and institutional capacity are likely to generate less corrupt polities with high levels of rule of law entrenchment. If legislatures can create an ombudsperson role, we can expect cleaner polities where officials respect the rule of law and do not abuse their delegated powers (Dolan and Bennett 2019; Fish and Kroenig 2009: 5–17; Stapenhurst et al. 2014: 290). The office of the ombudsperson would investigate graft, summon executive officials to testify about policy mistakes, open direct investigations on the executive and its officials, give approval for the appointment of ministers and heads of executive agencies, have oversight of agencies of coercion and change or amend the constitution to remedy laws that enabled malfeasance or create new ones to prevent it.

The means by which strong legislatures mitigate corruption and buttress the rule of law is based on their capacity to constrain the executive through institutional mechanisms directly available to them. In some legislatures, there is a lack of investigative committees or impeachment/removal procedures to deter the president or prime minister from engaging in corrupt activities or encourage them to abide by the rule of law. Legislatures that have more institutional capacity to perform the oversight mechanisms discussed previously are more likely to have a didactic effect in deterring other branches of government from engaging in any act of political malfeasance. Strong legislatures, in essence, generate priming effects that reduce governmental malfeasance. By impeaching the president, subjecting the prime minister to a vote of no-confidence, holding members of the legislature accountable through question time/constant plenary investigations and forming public account committees that perform national budgetary audits on a regular basis, legislatures lead by example and create an institutional environment conducive to prohibiting errant behaviour among public servants.

Some scholars may be sceptical of this linkage as controlling corruption involves prosecutorial authorities within national and local courts. However, judicial institutions and the courts have acquiesced and become mere rubber-stamping agents of the executive, especially in the developing world (Fish 2006; O'Donnell 1999). In many democracies, it is the legislatures that have a direct influence on the courts to prosecute members of the executive branch, which ultimately strengthens the rule of law (Pelizzo and Stapenhurst 2014). Judicial authorities have a constrained role in deterring corruption because judges in many countries are appointed by the executive as decreed by constitutions (Rose-Ackerman 2007). This collusion leaves it up to the legislature to expose the sycophantic predispositions of the courts to turn a blind eye to executive-level corruption. By exposing malfeasance through investigative blue-ribbon committees that are open to the media and public scrutiny, stronger legislatures that have institutional autonomy and specified powers to investigate the executive branch can compel the courts to perform their constitutional duty to prosecute graft-prone politicians and hold them accountable.

This leads to the formulation of the third hypothesis:

Hypothesis 3a: Strong legislatures are correlated with lower levels of state-level corruption.

Hypothesis 3b: Strong legislatures are correlated with higher levels of rule of law entrenchment.

Measuring legislative strength

Because there has been little examination of how legislatures affect political outcomes, scholars have assumed that other important institutional mechanisms are at play such that the power of legislatures are undermined by the strength of the executive branch, the effective number of institutionalized political parties, divergent electoral system design, whether a country has a parliamentary or presidential system, and the extent of political decentralization. Scholars in comparative politics have created a plethora of studies that seek systematically to operationalize measures for these institutional arrangements, neglecting to quantify the perceived power of global legislatures (Chernykh et al. 2017: 295). In response to this oversight, Fish and Kroenig (2009) conducted a global survey of political experts from more than 100 countries (within the years 2007–2009) regarding their perceptions of how much political power their respective legislature has, based on four important dimensions: (1) its influence over the executive; (2) its degree of institutional autonomy; (3) its specified powers codified in the constitution; and (4) its institutional capacity. Online Appendix 2 provides the list of questions given to the panel of experts from each country, from which a total of 32 specific powers are enumerated, representing the four above-mentioned dimensions. From the list of 32 specified powers, Fish and Kroenig (2009) construct a Parliamentary Power Index (PPI) score, which is derived as an aggregate mean score based on an additive index. Thus, the PPI is the total number of legislative powers a country has out of the 32 specific powers enumerated. Myanmar's Pyituh Hluttaw has the lowest PPI score at 0.00, while the German Bundestag has the highest at 0.84. Higher scores on this index indicate a legislative body that has more concentrated powers that can influence political outcomes. The advantage of this measure is to provide researchers with the ability to compare the perceived strength of each country's legislative capacity in a comprehensive and parsimonious fashion, allowing for the construction of cross-national questions on how legislatures influence policymaking. This measure is ‘the most ambitious effort to date to quantify the power of legislative bodies’ (Chernykh et al. 2017: 296). The PPI is utilized as the main independent variable in this study.

The PPI combines all power dimensions that a legislature can have – encompassing an attempt to operationalize legislative capacity as equally built on these facets. More specifically, by devising such a measure, scholars can make valid and reliable comparisons of the relative power of global legislatures in reference to the power of the executive structures cross-nationally (Wilson and Woldense 2019). Parsing the four dimensions out and disaggregating each feature's impact on effective governance would not make theoretical and empirical sense, as these power dimensions are interrelated and correlated.

The PPI has methodological shortcomings. The first concern is its use of an aggregation technique conflating a legislature's informal and formal powers. This reduces its efficacy as a valid and reliable independent or dependent variable because it is prone to generating endogenous research questions (Desposato 2012). Further, specific legislative powers may appear on the constitution, only to see that the legislature does not have the prerogative to practise or execute them and affect political outcomes because of limits imposed by other institutions. A flaw of the PPI measure is its use of an aggregation technique placing equal weighting on the four dimensions and the 32 specified powers enumerated. To address these weaknesses, Svitlana Chernykh et al. (2017) provide an alternative measure known as the Weighted Legislative Powers Score (WLPS), where they conduct their own survey of panel experts and reweight the 32 important attributes of legislative power by asking experts to rank them in order of importance from most important to least important. This was absent from the original aggregation technique of the PPI, which treated each of the 32 specific powers equally. The WLPS allows for ‘more nuanced distinctions about what is more or less important with regard to the substantive content of legislative power’, by up-weighting and down-weighting the 32 constituent components of the PPI (Chernykh et al. 2017: 296).

The WLPS is tapped as an alternative independent variable to ensure robustness of results. The WLPS variable rescales the PPI with a minimum score of 0.000 to a maximum score of 5.932. Both legislative strength measures are highly correlated, as depicted in the scatterplot presented in Online Appendix 1 with a Pearson's r value of r = 0.9976, indicating that both measures are essentially capturing the same concept of legislative strength.

Empirical strategy

Cases

The study employs a time-series cross-section analysis incorporating a global sample of 150 countries that have available data from the PPI and the World Bank's Worldwide Governance Indicators (World Bank 2020). To ensure that the results hold for a global sample and countries in the developing world, which may lead to divergent results as a function of how economic development may influence divergent governance outcomes, the study runs a reduced model restricted to low-income and middle-income countries as identified by the World Bank.²

Dependent variable

Effective democratic governance is conceptualized by the United Nations as a process of policymaking by governmental authorities that involves four important dimensions: (1) accountability; (2) effectiveness and efficiency; (3) openness and transparency; and (4) rule of law entrenchment and the absence of corruption (UN 2009). The underlying principle of effective democratic governance occurs when the public has the opportunity to participate in the political process without discrimination, through freedom of association, where the state's institutions generate tangible results reflecting public needs and demands using resources at their disposal with efficiency, where governments generate policies that are revealed to the public through open communications, and where governments apply legal frameworks and mechanisms fairly reflective of having incorruptible legal agencies (UN 2009). This definition is encapsulated by the World Bank's World Governance Indicators (WGI), widely used by scholars in capturing six important dimensions of governance, which include voice and accountability, governmental effectiveness, regulatory quality, control of corruption, the entrenchment of the rule of law, and political stability.³ The study employs all these dimensions except political stability, considering how the presence of threats against the state by external actors can be construed as a mechanism that can generate vertical or horizontal accountability (De Ferranti 2009). Such dimensions of good governance are based on expert surveys and aggregate global indicators. They are converted to an interval scale ranging from a minimum of −2.5 to a maximum of +2.5, where higher scores connote better governance, while lower scores reflect inferior governance (Kaufmann et al. 2009).

Controls

The first set of control variables pertains to other institutional variables that may cause variations on the effective governance measures. The variables include a dummy variable to represent parliamentarianism: a country that has a pure parliamentary system is categorized as 1, all else 0 (which includes presidential and mixed systems) (from Beck et al. 2001), and the number of veto-players, which identifies the number of independent branches of government with the ability to influence policy change in various countries on a yearly basis. To operationalize this, the study uses the veto-player data provided by the Political Constraints Index data set (Henisz 2018). Dummy variables connoting polities that utilize a proportional representation system and those that employ federalism are also utilized because they may influence variations in effective governance.⁴ The second set of controls involve the natural logarithms of the population and the gross national income (GNI) per capita of each state, considering that population pressures and lower levels of economic development may constrain the state from engaging in effective policymaking.⁵ The third set of controls incorporates levels of democracy using the Polity IV Index which measures the extent of institutional restraints on governmental institutions (from Center for Systemic Peace 2021) and ethnic fractionalization (from Alesina et al. 2003). These are employed because countries that have higher levels of democracy tend to be more accountable to their citizens (as part of the legitimization function), which enhances the quality of their bureaucracy and public services (Benz and Papadopoulos 2006), while countries that have ethnic cleavages are likely to suffer from substandard public goods provision (Alesina et al. 1999). Rafael La Porta et al. (1999) identified variables that may have an effect on the quality of government, including a legal tradition influenced by English common law (coded as 1; all else 0); a legal system influenced by a socialist legal tradition (coded as 1; all else 0); percentage of the population that is Protestant or Muslim in 1980; and the distance of each state from the equator (absolute value of a country's latitude from north to south with the equator as reference category). Lastly, interactive terms between the legislative strength indices and all institutional variables are included in all the models (not shown because of space constraints) as a control to ensure that the main effect of legislative strength is not attenuated or mediated by constitutional design, electoral system choice, veto-players or federalism.

Statistical estimation

The temporal frame of the pooled time-series cross-section analysis is from 1996 to 2016, constrained by data availability of the WGI, which is provided by the World Bank biannually from 1996 to 2002 and annually from 2003 to 2016. The panel data are arrayed based on country-time (year) format. Variance inflation factor scores of the explanatory variables do not exceed the threshold value of 3.00, indicating that collinearity issues are accounted for (Gujarati et al. 2009). Since the panel data are unbalanced, the statistical method employed is ordinary least squares (OLS) regression with panel-corrected standard errors (PCSE), which specifies a panel-specific AR1 correction to mitigate deficiencies associated with time-series modelling such as panel heteroskedasticity, serial autocorrelation in the error term, unbalanced country-year panels and non-stationarity (Beck and Katz 2011).

Addressing endogeneity

The high correlation between the PPI (measured in 2006) and the WLPS (measured in 2014) and the fact that they are based on expert surveys conducted in separate years (separated by eight years) connote that the perceived strength of legislatures is fairly constant over time. Both measures of legislative strength are not subject to high levels of variation across years as most are codified in founding constitutions of nation states and have become reified through time. These measures of legislative strength allow us empirically to test the effect of legislative strength on all measures of effective democratic governance that naturally attenuate endogeneity. This is because the independent variables of interest (legislative strength) temporally precede the dependent variables which are measured with a longer temporal time frame by the World Bank from 1996 to 2016.

To further mitigate endogeneity in the modelling, a lagged value of the dependent variables is included on the right-hand side of each regression equation. All time-variant explanatory variables are lagged at _(t − 1) since values within a specific year may be unduly influenced by its value in prior years (to rule out simultaneity). The summary statistics of all variables used in the analysis are provided in Online Appendix 4.

The study also presents (in Online Appendix 7) alternative models using a fixed effects vector decomposition (FEVD) procedure as a form of sensitivity analysis. FEVD analyses are utilized to address issues related to ameliorating reverse causality, the presence of time-invariant variables, and endogenous regressors (Plümper and Troeger 2011).

With these checks on endogeneity, we can be confident that the results flow from strong legislatures to effective governance rather than the reverse.

Results

The analysis begins by analysing the relationship between legislative strength and voice and accountability. As presented in Table 1, the OLS models with panel-corrected standard errors show that there is a strong positive relationship that is statistically significant at (p < 0.01), as demonstrated in both the global and developing states models. Even when controlling for the effect of other institutional variables, levels of democracy, economic development and other demographic factors, what we find is that legislative strength contributes to higher levels of voice accountability, which is conceptualized by the World Bank as citizenship empowerment, freedom of expression and association, and large-scale political accountability. These findings remain robust even when we utilize the reweighted WLPS measure that accounts for up-weighting and down-weighting of legislative powers deemed most and least important by experts.

Table 1 Legislative Strength and Voice and Accountability

Note: *** p < 0.01; ** p < 0.05; * p < 0.10 (two-tailed test). Panel-corrected standard errors in parentheses.

Do strong legislatures have an effect on promoting governmental effectiveness as well? Results in Table 2 confirm that an increase in legislative strength using both the PPI and WLPS index is associated with an increase in bureaucratic effectiveness that is statistically significant at (p < 0.01). The results are consistent even in the model with a reduced sample consisting of low-income and middle-income countries.

Table 2 Legislative Strength and Governmental Effectiveness

Note: *** p < 0.01; ** p < 0.05; * p < 0.10 (two-tailed test). Panel-corrected standard errors in parentheses.

Relatedly, do strong legislatures affect the formulation and implementation of policy that strengthen the protection of private property, contract enforcement and private-sector development? Results provided in Table 3 mimic a strong legislature's positive effect on governmental effectiveness, as shown previously. Across all model specifications involving the global models and models restricted to the developing world, we find the parameter coefficients of both the PPI and WLPS index are statistically different from 0, illustrating a robust, positive linear relationship between legislative strength and regulatory quality at (p < 0.01).

Table 3 Legislative Strength and Regulatory Quality

Note: *** p < 0.01; ** p < 0.05; * p < 0.10 (two-tailed test). Panel-corrected standard errors in parentheses.

We next delve into understanding how nation states with strong legislatures should have lower levels of government malfeasance and promote a culture of rule of law entrenchment. Results in Table 4, across all model specifications, convey that as legislative strength increases (measured by the PPI and WLPS index), the control of corruption also increases, reaching statistical significance at conventional levels. This positive relationship accords with the theoretical argument that strong legislatures have the institutional autonomy and oversight capacity to scrutinize the behaviour of executive-level officials or legislators that subsequently lessen collusion among elected officials and reduce governmental malfeasance. Relatedly, legislative strength (using the PPI and WLPS index) is highly positively correlated with rule of law entrenchment which is statistically significant at (p < 0.01) (in both models), as depicted in Table 5. This result corroborates case-study evidence that strong legislatures can subdue elected officials under the rule of law that can promote a culture of horizontal accountability (Barkan 2008).

Table 4 Legislative Strength and Control of Corruption

Note: *** p < 0.01; ** p < 0.05; * p < 0.10 (two-tailed test). Panel-corrected standard errors in parentheses.

Table 5 Legislative Strength and Rule of Law

Note: *** p < 0.01; ** p < 0.05; * p < 0.10 (two-tailed test). Panel-corrected standard errors in parentheses.

Marginal effect plots presented in Online Appendix 5 illustrate the strong, positive linear effect of legislative strength on all aspects of effective governance, while holding all control variables constant at their means and modes. It is important to note that none of the other interacted institutional variables with the PPI or WLPS reaches statistical significance at conventional levels. This demonstrates that legislative strength alone is a strong predictor of all the dimensions of effective democratic governance and is not weakened by its interaction with other institutional mechanisms ranging from constitutional and electoral system design, veto-players and federalism.

Sensitivity analysis and disaggregated effects

It is likely that the effect of the PPI on voice and accountability (quality of democracy) are endogenous – that is, contingent on regime type. To test if strong legislative institutions have an underlying role in promoting higher levels of voice and accountability in both democracies and non-democracies, a marginal effects plot was generated while holding constant all control variables at their means and modes and dropping Polity IV as a control variable. The results (see Online Appendix 8) show that legislative strength exerts the same positive effect on voice and accountability across both regime types that is statistically significant at conventional levels.

The FEVD analyses presented in Online Appendix 7 also generated results consistent with how strong legislatures are positively correlated with all the dimensions of effective governance, statistically significant at conventional levels.

Each of the four dimensions of the PPI is also used as an alternative independent variable to determine if it is correlated to effective governance measures. Results, presented in Online Appendix 9, demonstrate that all PPI dimensions (influence over the executive, degree of institutional autonomy, specified powers and institutional quality) have a statistically significant positive effect on all measures of good governance at conventional levels. However, it cannot be ascertained empirically which of these dimensions are more relevant in influencing effective governance because one dimension is interrelated with another and all are highly correlated. Thus, using the PPI that aggregates all four dimensions is a more valid and reliable proxy that approximates the generalized strength of global legislatures and its effect on good governance.

Discussion

The results have several implications. First, since legislatures may have the institutional capacity to enhance the quality of life for citizens through effective governance, and since effective governance promotes economic and political stability, international organizations and donor countries should prioritize legislative strengthening programmes as part of their chief objectives in allocating priority areas for financial assistance in the developing world (Kinyondo and Pelizzo 2013).

Despite this, many legislatures continue to lack professionalization, remain under-institutionalized and are incapable of exercising horizontal accountability and oversight responsibilities. In other words, they cannot check the concentrated powers of the executive and the bureaucracy that have increased in scope over time (Alabi 2009; Van Vliet 2014). In the absence of a well-functioning legislature, such polities are not likely to improve the status of their democratic systems pertaining to building durable democratic institutions. Weak legislatures cannot promote a democratic culture and build institutions that protect the civil liberties and political rights of citizens. Since most of these countries are nascent democracies, the likelihood of autocratization or creeping authoritarianism is likely, if legislative-strengthening programmes are neglected, abandoned or not fully appropriated (Kapstein and Converse 2008).

Second, even in consolidated democracies, strong legislatures could provide a bulwark for maintaining political accountability and governmental responsiveness that can sustain higher levels of political trust and political participation among critical citizens that have become apathetic and uninvolved in the political process (Norris 2011). This study finds that stronger legislatures promote higher levels of political voice. Accountability mechanisms provided by strong legislative bodies can strengthen the representativeness function of governments and enhance the performance of the government in addressing and meeting citizen demands, thus continually increasing the well-being and quality of life of global citizens, even in industrialized democracies.

Third, since strong legislatures may curb government malfeasance and strengthen the rule of law, there is a need to reorient the way legislative-strengthening programmes work. To date, many programmes to enhance the functions of national legislatures are working under the framework and guidance of the UNDP agenda (UNDP 2011), focusing mostly on legislative capacity-building, reforming the committee systems, fixing electoral system designs, developing institutionalized party systems and reorganizing leadership structures to ensure an inclusive environment in its representativeness function. Lacking from these programmes are institutional frameworks upon which legislatures can impose rigid checks on executive-level malfeasance, such as creating an independent office of an ombudsperson within the legislature, anti-corruption committees or subcommittees, stronger audit mechanisms, efficient public account committees and the standardization of plenary hearings that subject executive officials to questioning, interrogation or interpellation (Madue 2012). Such oversight capacity mechanisms, if strongly introduced and sustained within legislatures over time, can improve human development across states because financial losses and economic resources diverted to the pockets of rent-seeking elected officials can be obviated with a high degree of continued legislative oversight (Hudson and Wren 2007; Mungiu-Pippidi 2015).

In conclusion, this study is an attempt to bring back the significance and importance of the legislature in the realm of research in comparative politics, especially its ability to influence the effective functioning of governments. Future work should continue to construct and refine measures that approximate the strength of legislatures and theoretically and empirically unpack how they affect efficacious policymaking and improve the quality of life of global citizens.