«Democracy, Development and the Executive Presidency in Sri Lanka Rajesh Venugopal London School of Economics (LSE) r.venugopal © Rajesh ...»
DRAFT VERSION – FOR PRIVATE CIRCULATION ONLY
Democracy, Development and the Executive Presidency in Sri Lanka
London School of Economics (LSE)
© Rajesh Venugopal 2014. This is a draft paper for private circulation only. Please
do not cite without permission.
Institutions, constitutional change, Sri Lanka, democracy, presidentialism.
This paper examines the developmental causes and consequences of the shift from a parliamentary to a semi-presidential system in Sri Lanka in 1978, examining its provenance, rationale, and its unfolding trajectory. In doing so, it draws on a wide range of sources and a decade of research on Sri Lankan history and politics, and engages with debates on the role of institutions and democracy on development. It sets out an argument in brief that the executive presidency was born out of an elite impulse to create a more stable, centralised political structure to resist the welfarist electoral pressures that had taken hold in the postindependence period, and to pursue a market-driven model of economic growth. This strategy succeeded in its early years 1978-93, when presidents retained legislative control, maintained a strong personal commitment to market reforms, and cultivated alternative souces of legitimacy. In the absence of these factors, the presidency slipped into crisis in over 1994-2004 as resistance to elite-led projects of state reform mounted and as the president lost control of the legislature. Since 2005 the presidency has regained its power, but at the cost of abandoning its original rationale and function as a means to recalibrate elite/mass power relationship to facilitate elite-led reform agendas.
Acknowledgments: I would like to thank seminar and workshop participants at the University of Zurich and the London School of Economics for earlier feedback. All shortcomings are my own.
Democracy, Development and the Executive Presidency in Sri Lanka
INTRODUCTIONHow do political institutions affect economic development? Since the early 1990s, the theory and practice of economic development has been transformed by the idea that ‘institutions matter’ and that economic outcomes are best explained by the institutional environments that beget them. Consequently, a wide assortment of developmental pathologies have been traced back and reframed as institutional problems, whether of poor design, weak enforcement, non-transparency, accountability, over-centralisation, multiplicity, low capacity, weak legitimacy, or elite capture. The promotion of economic development is as a result widely understood to be a matter of engineering and enhancing an optimal and functional institutional regime of political rules, with a concomitant set of rewards and punishments.This paper contributes to that broad field of study by exploring the developmental causes and consequences of a major institutional change that took place in Sri Lankan development history: the shift from a parliamentary to a semi-presidential political system with the introduction of a powerful executive presidency in 1978. The semi-presidential system was introduced as part of a new constitution in 1978, and was the most significant change to the country’s political architecture since independence in 1948. Moreover, there was a distinctly developmentalist rationale that inspired it, and that it was brought into being to serve.
The main difference between a presidential and parliamentary system of government is in the relationship between the executive and the legislature. In parliamentary systems, the executive, including the prime minister and the cabinet, is composed of sitting members of the legislature. In a presidential system on the other hand, the executive is headed by a directly elected president, so that there is a greater separation of powers between the executive and legislature.
There is a significant debate in the political science and economics literature on the relative merits of presidential versus parliamentary systems for democratic resilience, government stability, and effective policy-making (Linz 1990, Shugart 1992, Cheiboub 2007, Elgie 2004).
Persson and Tabellini (2005:276) suggest that ‘presidentialism could lead to overall better policies in consolidated and solid democracies but not in more precarious democratic situations’. Cox and McCubbins (1991) argue that the separation of powers in a presidential system makes it less decisive in policy-formulation, but resistant to reversal and thus more resolute in carrying it forward through implementation. Keefer (2004) instead argues that policy outcomes are only weakly affected by constitutional choice.
Indeed, at first glance, the switch from a parliamentary to a semi-presidential system in Sri Lanka did bring about evident changes to development policy. The Westminster-style prime ministerial system of 1948-77 is widely characterised as a period when electoral populism led to the growth of unsustainable welfare spending, excessive state regulation of economic activity, and economic stagnation. In contrast, the switch to the Gaullist semipresidential system in 1978 gave birth to a more authoritarian politics and to a new era of market reforms, economic dynamism, and global re-integration.
1 Democracy, Development and the Executive Presidency in Sri Lanka But a closer inspection of the actual processes involved would reveal a far more challenging and complicated reality. Economic growth did increase in the years following the introduction of the executive presidency, but often in unexpected and counter-intuitive ways that were only indirectly connected to the new institutional system. The rapid growth and structural transformations that were evident through market reform-led growth in the south also occurred in parallel with the escalating civil war in the north-east, generating a schizophrenic mix of development amidst destruction. The market reform programme was itself no textbook shift from state to market: it was accompanied by a massive expansion of the state, first through large rural development schemes, and later through the fiscal impact of the expanding defence budget, each of which created knock-on effects within and beyond the economic sphere.
In other words, the superficial correlation between the presidential system and higher economic development provides an incomplete, if not entirely misleading answer, and leaves the most important questions un-addressed. More broadly, the study of political institutions and economic development struggles to address three key methodological problems. Firstly, there is the challenge of endogeneity, and understanding how to uncouple cause and effect: do institutions lead to development or is it vice versa? Secondly, while correlations abound, there is a need to better understand the actual processes and causal sequences. Thirdly, what are the other contextual variables at play that explain why the same institutions, even in the same country, can produce very divergent outcomes over time?
This paper seeks to address some of these prevailing challenges by taking a more actorcentric perspective of how institutional reform is rationalised, conceived and implemented, and then evaluates its outcome under that framework and set of criteria. That is, institutional change and the economic consequences that result are not cast within the confines of a causal relationship to isolate and test in itself, but as a political project to be approached from the perspective of a historically informed political sociology. This involves tracing the origins and rationale of the executive presidency in Sri Lanka, identifying its social constituency of support, and then explaining how this project unfolded in practice, and with what consequences.
In analytical terms, this trajectory is mapped out across the terrain of the broader relationship between democracy and development in Sri Lanka, and with particular reference to the ‘elite/mass discontinuity’, which James Manor described as the ‘principal political cleavage in the polity of Sri Lanka’ (Manor 1979:22). This paper takes Manor’s typology (written incidentally in the immediate aftermath of the 1978 constitution) as the point of departure to view how elites and masses (particularly with reference to the SinhalaBuddhist majority), positioned themselves on opposite sides of the main issues of political contention with respect to economic policy and also to the ethnic conflict.
The empirical basis for this paper is eclectic and diverse. It includes both primary and secondary evidence from a variety of different sources as warranted to advance, illustrate, and substantiate the narrative, but is largely based on an accumulated body of original
research and fieldwork conducted by the author on Sri Lankan politics and the economy from 2004-2012, based on interviews, documentary evidence, archival research and statistical data.
In brief, modern Sri Lankan constitutional history has four distinct periods: the Donoughmore constitution of 1931-47 that introduced universal franchise and a significant measure of self-government; the Soulbury constitution 1948-71, which created a Westminster style parliamentary system under dominion status at independence; the Republican constitution of 1971-77, which largely continued the basic domestic political architecture of Soulbury, but which ended dominion status; the 1978 Gaullist constitution which introduced an executive president and proportional representation-based voting for the legislature.
The academic literature on development as well as the guiding wisdom during the colonial period was that development has to precede democracy: stable democracies are only tenable at later stages of development with higher levels of income and education. This is the thrust not only in Seymour Lipsett’s classic work on the subject, (Lipset 1959) but also in Adam Przeworski, who comes to the same conclusion from an opposite direction of causality. Whereas Lipsett essentially argues that the poor are ill-suited to democracy and vulnerable to demagogic extremists, Przeworski argues that it is the rich who will find democracy intolerable because of the threat it poses to their wealth (Przeworski 2008).
Indeed, Przeworski et al (1996) attach a statistical probability of 0.12 for democratic survival in a poor country with per capita income of less than $1000 – implying that the expected lifetime of a poor democracy is 8.5 years.
As such, theory holds that premature democratization in poor, under-developed countries, such as late-colonial Ceylon would cause either democracy or development to suffer. Such countries would either revert to non-democratic authoritarian regimes (in form if not in substance), or else suffer extended periods of retarded and distorted development. Or else, they could chaotically zig-zag through a half-way system where weak democratic institutions and weak developmental outcomes reproduce one another. In contrast, the experience of successful late-developing countries in East Asia such as South Korea and Taiwan is illustrative: both remained under authoritarian regimes during the period of their industrialisation, and did not democratize until the 1980s, by which time they had already achieved a substantial measure of prosperity.
This mainstream consensus has of course endured a vibrant set of critiques and alternate theories of the sequence from economic development to democracy (Moore 1966, Evans 1979, O’Donnell 1973, Nelson and Huntington 1986). Nevertheless, it remains significant not because of it has greater explanatory power or empirical validity as such, but because it has wide currency and are influence on the ground. As such it has relevance not just for its analytical content but because it is a component part of that reality under study.
The argument in brief is that Sri Lanka’s executive presidency was born out of an elite impulse to create a more stable, centralised and authoritarian political structure that would overcome and reverse the negative economic effects of a populist electoral democracy. It 3 Democracy, Development and the Executive Presidency in Sri Lanka would they hoped, help revive economic growth under a market-driven development regime. This project succeeded (on its own terms) in its early years, when the presidencies of J.R. Jayewardene (1978-88) and Ranasinghe Premadasa (1989-93) retained legislative control and maintained a strong personal commitment to market reforms. It later struggled and went into crisis under the Chandrika Kumaratunga presidency (1995-2005) as resistance mounted from above and below. The presidency itself was rescued from crisis by Mahinda Rajapakse presidency (2005-2015), but at the cost of rejecting the project and rationale that gave rise to it and by embracing the electoral populism that it was created to resist.
TASTING THE FRUIT BEFORE PLANTING THE TREE: 1948-77 The internal debate on democracy and development needs to be traced back to the colonial government reforms of the late-1920s, and the hearings of the Donoughmore Commission that was entrusted with making recommendations for a new constitution.
Political reforms had since the 1880s gradually expanded the quantity and quality of native representation within the colonial government, and the nascent Ceylonese elite that had been thus drawn in expected and lobbied for further such gradual reforms. However, the Donoughmore Commission, which arrived in Ceylon in 1927 had more ambitious and radical reforms in mind. They proposed a significant expansion in self-government but linked this to an even more significant expansion of the franchise – hitherto restricted to men of education and property – to all men and women aged 21 and over.