Continuity and change in Cuban policy making during the 21st century

(1)

Continuity and change in Cuban policy

making

during the 21

st

_century

An analysis of the agricultural reforms between 2007 and

2014

Peter Wiersma, s4051637

(2)

Supervisor: Dr. A. Wigger 27 March 2017

(3)

(4)

Abstract

Since 2007 far-reaching socio-economic reforms have been implemented in Cuba, which seem to indicate a transformation away from direct state intervention towards a more free-market attitude in areas like agriculture, housing and the financial sector. Several scholars have suggested that these reforms need to be understood as a ‘socialist transition’, a transition in which the Cuban government tries to maintain (economic) stability by granting more liberties to the Cuban citizens, while at the same time ensuring the continuity of a socialist model in Cuba. Policies aimed at the development of sustainable agriculture in the 1990s have marked a starting point of these changes. Land reforms in 2008 and 2012 suggest a radical shift in this domain; usufruct laws allowed for individual farming opportunities and more market-based production and distribution of agricultural products. Whether this shift represents a gradual and thus incremental institutional change within the existing system, or whether it marks the ascendancy of a complete turnaround, with a commitment to capitalist organization of the economic realm remains to be seen in the future. This thesis explains the political processes that have led to the shift from state regulation to a system with more liberties in the agricultural sector, as embodied in the policy reforms 2011 Lineamientos – guidelines – adopted by Cuba’s Sixth Party Congress. It sets out to explain this by assessing two models of institutional change, aiming to extend the institutionalist debate on the discussion about endogenous and exogenous influences on policy change.

Key words: Institutionalism – Cuba – Cuban agriculture – Cuban transition –

(5)

Introduction

“To such strategic tasks as food production, which as we have indicated is a matter of national security, we shall continue attracting the highest possible number of people through all the existing forms of property but in an orderly fashion.”

Raúl Castro,

2009

This excerpt was taken from Raúl Castro’s speech to Cuba’s National Assembly that took place on the 1st_{of August 2009. It refers to Cuba’s food shortages, as} food imports – accounting for eighty percent of Cuban meals – had dropped by a third (Latin America Herald Tribune, 2009). While Castro did not state it directly in his speech, the message conveyed that food production had to be seen as an urgent matter. Prior to his speech, a series of hurricanes had struck Cuba in 2008, damaging important infrastructure and destroying crops. They marked a starting point of efforts that were undertaken to reform the economic model and boost Cuban food production. This is referred to as actualización, or ‘update’ (Font & Riobo, 2013: 3).

In 2011 Cuba’s Sixth Congress of the Communist Party of Cuba (PCC) approved the so-called Lineamientos or ‘guidelines’. The Lineamientos display a much wider range of changes to be made in Cuba’s economic development than had previously been the case. They consist of measures in the economic, financial, social and agricultural field, intended to “update the Cuban economic model” (PCC, 2011: 6). The updating process began in the agricultural sector in 2008. Then it became possible for individuals and cooperatives to request idle land from the Cuban government for farming purposes. Whereas previously state-controlled farms cultivated the majority of lands, in recent years we see a significant increase of self-employed farmers and cooperatives in Cuba’s agricultural sector (Mesa-Lago, 2014: 52).

The Lineamientos introduced additional measures, which made it possible to sell agricultural products directly to the tourism industry. Previously, this had only been possible on special farmers’ markets. Linkages between producers and (end) consumers were also simplified; it became possible for small farmers to open bank accounts to receive direct payments from their customers in the

(8)

tourism industry (Gómez, 2014: 34). A related measure included the abolishment of the dual currency system by 2016 and the accessibility of credit for self-employed persons (Alejandro & Villanueva, 2014: 85). Possibilities like these extended not only to the agricultural sector, but also to other sectors, such as the construction and transport sector.

The fall of the Soviet Union entailed a major shift for world politics – in reality – but also in political science itself, renewing the debate on (dis)continuity of state behaviour. This applies to Cuba as well. Perhaps the most prominent question is why the collapse of the Soviet Union was not followed by a collapse of the political system in Cuba, as was the case in many Eastern-European countries. State socialism in Cuba persisted and developed on its own. Contrary to full Soviet collectivization, in Cuba there has always been land ownership, even after the revolution. Cuban policymaking after 2007 clearly deviates from the socialist principles of the Cuban political system we have seen until then. Policy objectives seem to be ‘contradictory’ with other policies, or socialist objectives. The changes in the agricultural sector display this ambiguity in several respects. To give an illustration: individual farmers are now allowed to build houses and other infrastructure on land that is owned by the government. However, it does not become their property.

Similar contradictions are found in the 2011 Lineamientos; although it is recognized that the economic and social model should be ‘updated’ to ensure economic prosperity, some of the measures seem to go against Cuba’s longstanding socialist tradition. On the one hand state-planned economy and non-ownership on private property are part of the socialist framework in which the guidelines have been set up. On the other hand, their main features – to improve production outputs and increase autonomy – display for the long-term a program that includes decentralization and allows individuals for more market-based production and distribution (Diaz, 2014: 6). This tendency towards less state regulation and more flexibility for producers and consumers appears to be part of ongoing changes that favor deregulation of the Cuban economy. Over the last ten years reforms in the agricultural sector seem to follow this trend. Nonetheless, it remains to be seen if this transition represents an absolute turnaround in Cuban policymaking.

Several research projects have been devoted to this phenomenon. For example: The University of Miami has a special institute called Cuba Transition Project

(9)

(CTP). Another major institute is the ‘Cuba Project’ by the Bildner Center and City University of New York. Both projects cover a wide range of studies to political, social and cultural change, ranging from the role of the state in Cuban civil society to US-Cuban relations. Economic aspects, such as market reforms, international trade relations and monetary policy are also widely covered (Feinberg & Piccone, 2014). However, the agricultural reforms have thus far not been researched in relation to Cuba’s political institutions and its specific policy process. Understanding this relation may provide insight into what makes the transition possible and the way it is to be interpreted in the light of Cuban socialism. As such it is the nature of this study.

Within the field of political science, there are various theories that seek to explain regulatory shifts like those in Cuba. Shifts imply change or discontinuity. The academic literature differs on how change comes about, what drives it and which factors explain continuity or discontinuity. Within mainstream International Relations (IR) theories there are several schools one can think of, such as realism, liberalism and social constructivism. These approaches consider the dynamics between the domestic and international level in terms of (world) systems, power relations and material interests, or through social constructs (Nye & Keohane: 1987; Waltz: 1979; Ruggie: 1998; Wendt: 1999). For instance: the role ideas play in a state’s identity, or how (international) norms on political behaviour come to be. Theorizing state behaviour and what constrains it, is often key in their explanations for change or continuity in politics.

However, to interpret Cuba’s agricultural policies, we need to go beyond the scope of state behaviour. While state behaviour may provide comprehensive explanations for political outcomes on state level, this alone is insufficient to explain why there are contradictions in Cuba’s policymaking. Understanding how the Cuban state – or any state – behaves is a first logical step to make. Nonetheless, with an understanding of what a state represents and how it functions, we do not touch upon what makes state behaviour contradictive. For this reason, we need to relate it to the political environment, be it internationally or domestically. Both levels are relevant to the Cuban case; the US trade embargo or the level of food production may be equally important to explain Cuba’s agricultural policies. Since the policy process itself also seems to be key here, we need to understand which mechanisms are essential to policy change and how these connect to Cuba’s internal and external political

(10)

environment. Within the political and social science debate on change, recent schools of institutionalism aim to address such relations.

Though early 20th_{century institutionalism only concerned itself with the} description of state structures formalized into institutions, rule and law, in the second half of the 20th_{century, the field developed a broader notion of} institutions that was not solely confined to the formalization of government (Bell, 2003: 4). Institutions could also be seen as an explicitation of belief systems on the functionings of society, shaped by ideology, social norms, or ideas (Hall & Taylor, 1996: 946). This means that societal actors such as interest groups, media or political parties can be included in institutionalist frameworks. The autonomy of the state vis-à-vis society and institutions, and institutional behaviour, became principal objects of research in institutionalist studies (Bell, 2003: 7). Varying assumptions on rationality, social interaction, the influence of culture and historical development, have led to different perspectives on institutional change.

Recent approaches to institutionalism disagree which mechanisms affect the behaviour of political actors through institutions and vice versa. Such approaches are commonly referred to as ‘new’ institutionalist approaches (Bell, 2003; Hall & Taylor, 1996). The discussion involves whether processes of change are mainly driven by an exogenous or endogenous dimension. To illustrate: if the research unit is the ‘state’ and the aim is to explain the behaviour of a state, explanatory factors outside the state are exogenous, while explanatory factors inside the state are endogenous. Institutional change is often considered to be prompted by explanatory factors in the exogenous dimension, while no, or little attention is given to endogenous explanations (Mahoney & Thelen, 2010: 5-7). Criticism revolves around emphasizing one dimension too much over the other, while it is insufficiently theorized why such choices are made. Clarifying this theoretical discussion is central to this thesis.

Two historical institutionalist models will be used to analyze the Cuban agricultural policies and to trace their origins. Historical institutionalism views the outcomes of decision-making primarily as a result of power relations between institutions and social groups. The causal effects of specific historical events, so-called critical junctures, are regarded to be essential for institutional development (Hall & Taylor, 1996: 945). It seems logical to address specific events that may have affected – or even caused – the course of recent Cuban

(11)

policymaking. However, the factor time should not be overlooked. Amenta (2003: 114) states the need for more argumentation by institutional scholars when “timing and sequence matter in the explanation of outcomes.” The question arises how we assign value to events that have already occurred and may, or may not, have had an impact on the (ongoing) policy process.

Although both institutionalist models pay attention to continuity in the policy process, the first model by Peter Hall (1993) emphasizes the possibilities for discontinuity. It is based on Hall’s theoretisation of the policy paradigm. A policy paradigm is a framework of ideas and norms on political reality reflected by policymakers (ibid.: 279). Hall’s model employs an ideational perspective to policy. Instead of focusing on institutional properties that make policy change possible, policymaking is theorized as a learning process taking place in the institutional environment.

The second model is by Mahoney & Thelen (2010) and theorizes how institutions structure policy by addressing the gap between the rule and its implementation. They have developed a framework that is – presumably – better able to account for both exogenous and endogenous change than previous models of change, the so-called Gradual Change Framework (GCF) (ibid.: 5). It is a causal model explaining how policy change might happen on a continuous basis, without an emphasis on specific dimensions. Mahoney & Thelen aim to describe the properties of institutions that make them vulnerable to change, either endogenously or exogenously.

Both models are employed to analyze the Cuban case, namely, to compare which explanatory factors – endogenous or exogenous – have been essential to the Cuban policy process. Involving the broader political environment – internally and externally – brings more clarity to the question how dimensions play a role in the Cuban policy process. However, it is not the aim to compare the models on their respective explanatory value in this matter. The reason for this is twofold: first, the models are not mutually exclusive to one another and could be complementary. Second, each model employs different mechanisms to regard institutional change. Nonetheless, they may provide different insights into the dimensions that play a role in the Cuban policy process. Such nuances may be essential to understand how Cuban policy is structured and may provide a more generalizable framework for policy change. Additionally, it may contribute to an

(12)

understanding of why and when state structures are susceptible to institutional changes.

The following research question is derived from the inconsistencies that are observed in Cuban policy:

What explains the changes in Cuba’s agricultural policies since 2007 towards the introduction of market-based elements in the agricultural sector with less state regulation, more flexibility and economic liberties?

To tackle the research question, a theoretical overview of (historical) institutionalist theory will be presented in the first chapter, specifically with how change is regarded. The emphasis on the exogenous dimension will be discussed, as well as its relation to the degree of autonomy of the state. The models by Hall and Mahoney & Thelen will be explained in further detail, including the content of the empirical indicators they describe. In addition, hypotheses are derived from the main explanatory variables of both models, specified to Cuba’s agricultural policy transition. Chapter two, the methodological section, first presents the used research methods, sources and data. The section then outlines the nature of the agricultural policy change in the explanandum. This is followed by the operationalization, which discusses the selection of the Cuban institutions, presents the concrete operationalizations of the endogenous and exogenous factors and empirical indicators of the models. In chapter three, the policy measures that have been taken between 2007-2014 are outlined into detail, followed by a discussion of the most notable changes they reflect. Chapter four, starts with an overview of the exogenous and endogenous factors that have influenced the Cuban policy process and are relevant to the application of both models. Next, the empirical indicators for the institutional types of change of each model are analyzed in relation to endogenous and exogenous factors. The findings will be presented with regard to the various hypotheses of both models. In the final chapter the conclusion is drawn, also discussing the limitations of the study and suggestions for further research.

(13)

Chapter 1 Theoretical framework

The impetus of this chapter is to present a clear theoretical basis to be able to interpret the agricultural policy changes that have taken place in Cuba since 2007. The historical institutionalist models by Hall (1993) and Mahoney & Thelen (2010) will be applied to this case. First, a brief theoretical background of the institutionalist field describes the context of the debate on institutional change relevant to the models. For this purpose, the origins of institutionalism will be discussed, as well as its relation to recent perspectives on the state and state autonomy. The relation of the state vis-à-vis institutions and society will be touched upon. This is the starting point for a short discussion of general assumptions historical institutionalist theory holds relevant to institutional change. These form an important basis for the two models and have implications for the exogenous and endogenous dimension of institutions, which will be examined as well.

Second in this chapter is a detailed overview of the theoretical models, describing which mechanisms and empirical indicators are used to analyze institutional change. An explanation of Hall’s concept of policy paradigms and paradigmatic change will be given, followed by an account of gradual change as presented in the framework by Mahoney & Thelen. The respective hypotheses on the institutional types of change – 3rd_{order change and gradual change – are} deduced after the discussion of each model.

Continuing, a comparison of the models is made; this includes a visualization of the described mechanisms and the indicators that will be used for the analysis. This provides insight into different dimensions of the policy process that are emphasized in the models, but may also point to specific nuances that are important for the analysis of the Cuban case. The aim is to identify similar empirical propositions and juxtapose the propositions that fundamentally differ.

(14)

For explaining political and societal developments, perceptions on the ‘state’, have been an important starting point for social and political sciences, specifically for the field of institutionalism.

“The state is considered the sole source of the ‘right’ to use violence. Hence, ‘politics’ for us means striving to share power or striving to influence the distribution of power, either among states or among groups within a state.” (Weber, 1921: 7)1

This statement is by 20th_{century sociologist Max Weber and was published in} Politik als Beruf, or ‘Politics as a Vocation’. Weber (ibid.: 12) posits the idea that

the modern state dominates the means of power and therefore, the means of political organization. In his understanding, political organization involves every act of power affecting how a state is organized. Weber’s perspective would become important for the discussion whether structural characteristics of the state could affect the policy process and – more generally – how political outcomes on state-level could be explained in relation to a state’s domestic (society) or international environment (states or international organizations). After World War II, so-called pluralist and Marxist approaches were dominant in the field of political science and international relations theory. Pluralist approaches did not consider the state as highly influential or consequential for political life. Instead, their focus was on the variety of political actors – elected officials, citizens or specific interest groups – that were able to influence political life.2_{Pluralists emphasized that actor’s opposing interests were} essential to political outcomes. Their argumentation was based on the relation between the available (political) resources and the ability of actors to access them. The state was only considered as a political platform or tool, enabling such relations. Thus, the result of public policy – and the ability to influence it – was seen as a result of different interests and resources between actors (Krasner, 1984: 227).

Marxists contested the pluralist idea of attributing political agency to actors best able to access political means. Where pluralists did not give attention to the state as such, Marxists argued that the state reflected class society. Their main argument to support this claim is that the state serves the interests of the elite class, whose aim is to maintain the status quo of capitalist society. A 1985 article by Theda Skocpol criticized Marxist and pluralist views of the state being merely a stage for political conflict. This criticism initiated the debate on

(15)

state-institutionalism and comparative case research. It involved theorizing institutional patterns that could explain the variation of political outcomes across cases (Amenta, 2005: 98-99; Steinmo, 2008: 158).

Responding to Marxist and pluralist views3_{, institutionalist scholars} ‘reintroduced’ the Weberian idea of the state and made it their own; “States were sets of organizations in some ways like other organizations but with unique political functions, missions, responsibilities and roles.” (Amenta, 2005: 99) They can be considered as actors with specific organizational structures – institutions – that centralize political authority in a given society.

Samuel Huntington, a 20th_{century political scientist, known for his influential} work The Clash of Civilizations (1993) published in 1965 an article on political development. Therein, institutions are defined as follows: “Institutions are stable, valued, recurring patterns of behaviour.” (1965: 394) In this view institutions are behavioural patterns determined by specific rules and norms, either formal or informal. Huntington believes that the level of institutionalization depends on how well (political) organizations are able to adapt and persist in a changing environment. The more adaptable an organization is, the more it is institutionalized in its given environment. Also, the broader the organization’s influence on its environment, the more it is institutionalized. Arguably, both conditions hold for the position of the state today; in modern day society it still exists as the highest form of political organization. It could be said that the state is institutionalized in society itself, but also among states, in the international environment. But what happens to these institutional patterns if there are changes in the state’s environment? Institutionalist theory attempts to describe how (state-)institutions constrain formal and informal behaviour of political actors, shape political outcomes and persist or change over time. They address what happens to institutional patterns if changes in the political environment (of states) occur. This relates to what extent institutions matter with regard to the autonomy of states and their political environment. Theoretical conceptions are generally concerned with two central issues. The first issue is about the extent of the autonomy of the state. That is: “can the state formulate and implement its preferences?” (Krasner, 1984: 224) The second issue is the level of coherency between the state and its environment, domestically or internationally. It concerns the question whether state structures influence changes in the environment and vice-versa. For

(16)

instance, Skocpol (1992) argued that state formation in the United States – the institutionalized federal system – had a profound impact on how political parties and interest groups dealt with (the introduction of) social policies in America. Among contemporary institutionalists4_{politics is often conceptualized in} different ways, as a process of power relations and (rational) preferences of political actors, a social learning process, or as an expression of culture and social norms (Hall & Taylor, 1996). Despite these conceptual differences, there is a similar understanding to be pointed out; there is agreement that there are established patterns of behaviour in states – institutions – which seem to be influential for political outcomes, and thus, for change or continuity. However, the main disagreement concerns how and when these patterns of behaviour manifest and to which extent. That is, do institutional structures – in states, political parties, international organizations, interest groups, etc. – mainly determine political behaviour, or does this also depend on the free choice of individuals? Or perhaps on combinations of both and on specific moments?5 Theorization of the endogenous and exogenous dimension of institutions is necessary to answer such questions. Distinguishing the internal from the external sources of change provides insight into how (state-)institutions, or individuals in institutions, direct institutional outcomes – in the form of rules, procedures, policy – in specific circumstances. Furthermore, understanding endogenous change in institutions is essential to explain institutional outcomes that have developed over time. The next section discusses the historical institutionalist view on this matter, since this approach provides the basic framework for the two models used to investigate the Cuban case. The emphasis lies with how political change, or continuity, is explained in relation to the

endogenous and exogenous dimension of institutions.

1.2 The theoretical approach of historical institutionalism

Like other institutionalist approaches, historical institutionalists treat states mainly as other organizations; they argue that (state-)institutions can be instrumental in shaping the context and roles in which political actors must behave in (Hall, 1986; March & Olsen, 1989; North, 1990). Historical institutionalism can be characterized by the view that political change is largely determined through the contextual features of the past. It considers politics to

(17)

be about asymmetrical power relations between institutions (Hall & Taylor, 1996: 939).

Institutions raise resource considerations reflecting a ‘struggle’ for power in politics. Historical institutionalists hold a distinctive understanding on institutional development, whereby an emphasis is put on specific historical events, or so-called critical junctures (Mahoney & Thelen, 2010: 6). For this reason, the perspective is called ‘path dependent’. This “refers to situations in which previous decisions set up the boundaries for future development.” (Ekelund, 2013: 104) An inductive logic is followed; this means that the causal effects of institutions on the policy process are emphasized when explaining institutional development and political outcomes. While historical institutionalists mostly emphasize power relations in this regard, they also believe that ideas and worldviews may play an important role in this process (Hall & Taylor, 1996: 940). Institutions can be regarded as a reflection of the dominant social culture, in terms of shared norms, values and identities (Ekelund, 2013: 103).

Following Blyth (2002: 2-3) the logic of historical institutionalism is consistent with models of punctuated equilibrium.6_{This means that change only occurs} within a brief timeframe during a moment of ‘punctuation’, after which a new institutional equilibrium is established. Thus, the source of change is exogenous to the existing institutional equilibrium and leads to a situation that is ‘vulnerable’ to change (punctuation); in turn this leads to a new institutional equilibrium. Several scholars have argued that moments of punctuation are likely to be preceded by a political or economic crisis (Blyth, 2002; Krasner, 1984; Wigger & Buch-Hansen, 2014).

Assigning causal relevance to such punctuated moments in relation to the later policy process and policy outcomes requires proper theorization. An illustration: to explain why the trade embargo of the US against Cuba still holds, one could look at the Cuba crisis of 1962 as a moment of ‘punctuation’. This crisis set in motion events that would result in a trade embargo put into effect by the US Congress. This was a clear turning point for American-Cuban trade relations and defining for later Cuban and American policy making. If we would want to explain how it has been defining for the trade embargo today, it would be much more complex to explain it simply as a ‘causal effect’ of the Cuba crisis. Instead, one would have to assess how and why American policy making on the trade

(18)

embargo has remained unchanged since the Cuba crisis. For instance, by looking at the power relations between Democrats and Republicans, or through the political ideas that have driven the policy course since then.

The point is that it would be easy to ascribe the Cuba crisis ex ante as the primary reason for why the trade embargo still exists, without proper theorization on timing and results of a crisis. The emphasis on historical (and causal) relevance of punctuated moments – critical junctures – as part of the exogenous dimension, may disguise how, endogenously, institutional outcomes have changed or remained constant over time.

The lack of theorization on the exogenous dimension is the most important criticism on historical institutionalism, which is expressed by various institutional scholars (Amenta, 2005; Blyth, 2002; Ekelund, 2013; Hall & Taylor, 1996; Mahoney & Thelen, 2010). Blyth (2002: 4-5) argues in this respect that we should pay attention to the role ideas play, since they “reduce uncertainty by defining a given moment of crisis and project the institutional forms that arguably will resolve it.” It draws attention to another issue relating to the endogenous dimension. If critical junctures are so important for institutional change, what are then the internal processes that make institutional arrangements – such as in the state – vulnerable to critical junctures? These processes are often neglected.

The models by Hall (1993) and Mahoney & Thelen (2010) attempt to address the lack of theorization in historical institutionalism. Most notable differences between their models are the manner in which they incorporate – to a greater or lesser extent – endogenous and exogenous indicators to describe processes of institutional change. The respective models are presented in the next section, followed by a brief discussion on their similarities and differences. It explains when they expect institutional change to be radical, and whether or not radical institutional change is per definition sudden, or can also occur gradually.

1.3 Theorizing institutional change: two historical institutionalist

approaches

1.3.1 Model one – Hall

(19)

The model by Hall (1993) aims to address the ideational dimension of the policymaking process leading to policy change. Policymakers have ideas on the purposes of policies, in terms of what they should do and how they should be implemented. According to Hall, such views on policies are shaped by the political environment policymakers operate in, a so-called ‘policy paradigm’. A policy paradigm “(…) specifies not only the goals of policy and the kind of instruments that can be used to attain them, but also the very nature of the problems they are meant to be addressing.” (ibid.: 279) These paradigms contain ideas and norms on political reality, which are reflected by policymakers in their policies.

Hall has derived the policy paradigm from Thomas Kuhn’s conception of scientific paradigms in The Structure of Scientific Revolutions (1962). According to Kuhn science progresses in a distinctive manner; during so-called periods of ‘normal science’ a specific scientific ‘paradigm’ is followed. This is a framework of ideas and assumptions scientists adopt on science, in terms of scientific practice and generally accepted knowledge and experience (ibid: 11). In a period of normal science, scientists aim to sustain the existing paradigm by expanding its empirical content and making it more specific. However, as a result, it becomes likely that anomalies are discovered that contradict the paradigm. In such instances, a paradigm may be adjusted or revised, without being rejected. Another possibility is that the anomaly results in a crisis, because it entails a revolutionary breakthrough incompatible with the existing paradigm. In turn, a competing paradigm may arise that – when generally accepted – becomes dominant (ibid: 85-86). Competing paradigms are unable to communicate with one another, since they require a different mind-set to ‘acknowledge’ the other perspective. For this reason, switching paradigms is a matter of choice between incompatible perspectives, also commonly referred to as a ‘paradigm shift’. Kuhn (ibid: 94) compares this process to political revolutions; the aim is to fundamentally change existing political institutions that hold standards and values on society incompatible with the revolution. Hall argues that the policy process develops in a similar way as Kuhn’s scientific paradigms. Policy paradigms can change but maintain their broad appearances, but they can also be replaced by an entirely new paradigm. However, a significant difference is considered: while policy paradigms can be replaced, the process leading up to a shift is political of nature, rather than scientific. This

(20)

means that choices to switch paradigms are politically motivated, unlike the pure scientific considerations that make scientists switch paradigms. This political aspect has three implications for Hall’s model of paradigm shifts;

First, Hall believes a ‘political’ paradigm shift follows a logic that is based on the power configuration between groups of actors in the institutional setting. One can think of administrators, politicians, media and interest groups. Access to “ancillary resources” and “exogenous factors” affect the relative power they have to change the paradigm. A switch of paradigms will depend on how well these actors are able to impose the other paradigm, depending on their “positional advantages” in the broader institutional setting (1993: 280). To illustrate: take a policy paradigm based on a strictly regulated state economy. Administrators will likely only implement policies through the administrative means they have to regulate the economy, while politicians have the authority to come up with new policy ideas. As such, the politician has a better position to impose a paradigm than the administrator.

Second, politicians decide who is authoritative in a certain policy area. In case of ambiguity between experts on policy issues, competing politicians have to decide who holds more authority. It is likely a paradigm shift is preceded by a debate on who holds authority on the policy in question, rather than a focus on the policy issue itself (ibid.).

Third, a shift is likely to be preceded by policy experiments and failures. The analogy with the scientific paradigms is again applicable; the adjustment of policies proceeds with the aim to sustain the status quo of the current paradigm. However, while such attempts may be undertaken with a specific goal in mind, they may also include random attempts to revise policy. Policy experiments and failures are expected in cases where a ‘weak’ policy paradigm exists. If the attempts are unsuccessful, policy failures may weaken the policy paradigm further and undermine its overall coherency (ibid.).

Institutional types of change

In model one policy change is characterized into hierarchic types, 1st_{, 2}nd_{and 3}rd order change. Policy change is observed in “the instruments settings, the instruments themselves and the hierarchy of goals behind policy.” (ibid.: 278-279) The hierarchy of goals is the priority assigned to different policy goals – for

(21)

example – decreasing food scarcity has more priority than decreasing unemployment rates. The instruments refer to material or political resources to achieve the policy goal; for instance, the cultivation of agricultural lands, or food imports. The setting refers to the precise realization of the instruments; training more farmers, or close trade agreements.

Instances of 1st_{and 2}nd_{order changes are considered adjustments of the policy} paradigm that keep the overall structure intact. Examples of 1st_{order changes} are cases wherein the setting of the policy instruments changes, but the hierarchy of goals and the instruments remain the same. This differs from 2nd order changes, that correspond with the use of different policy instruments, while maintaining the hierarchy of goals. If changes simultaneously occur in the three variables, institutional change is radical. This last type – 3rd_{order change –} is a ‘paradigm shift’, which essentially is a turnaround of the existing policy discourse.

Hall (ibid.: 278) believes that 3rd_{order change can develop through the} mechanism of social learning: “(…) a deliberate attempt to adjust the goals or techniques of policy in response to past experience and new information.” The three variables behind social learning – ‘goals that guide policy’, ‘policy instruments and the ‘setting of policy instruments’ – are meant to disaggregate the ideas that drive this mechanism. The main thought entails that policy objectives and the manner of implementation of the policy, tells us something about how ideas behind policy at time-1 have changed compared to policy at time-0.

The core of social learning is that policy will likely be the response to previous policy, whereby policymakers learn from the experiences of previous policy and may act to adjust it or keep it intact. 7_{This may work in two ways: first, similar} policy instruments of previous policies are utilized because they are known, or are seen as an appropriate response. As such, continuity of policy is to be expected. Second, responses by policymakers are driven by the consequences previous policies have had for the socio-economic and political environment. Policymakers learn of these consequences and adjust their responses accordingly. Thus, policy change is more likely to be expected.

The question is under which circumstances these two aspects change the policy paradigm. If a paradigm is vulnerable – for example, because of several policy failures – it seems plausible that policymakers pay more attention to the

(22)

consequences of endogenous factors in the socio-economic environment that undermine the policy paradigm. This seems logical, since they are able to change the policies that may have caused these consequences. These changes can be reflected by an adjustment of policy instruments, or – more far-going – a change of policy goals.

On the other hand, Hall (ibid.: 291) expects that “outside pressures” make policymakers more susceptible to undertake an attempt to deal with consequences of earlier policy. Otherwise, if no actions are undertaken, exogenous pressures can contribute to further weakening of the policy paradigm (such as more failures). In turn, this would make it more likely that 3rd_order change takes place.

Hypotheses on 3rd_{order change}

In the Cuban case, attention will be given to 3rd_{order change. The reason for} this is as follows. Instances of 1st_{and 2}nd_{order change are most likely} incremental changes of the policy paradigm. Hall points out that 1st_{and 2}nd order changes are not necessarily followed by 3rd_{order change, since they may} be used to support the existing policy paradigm (ibid.: 280). However, 3rd_order change supposes something else; change occurs simultaneously in the three variables of policy. The reforms in the agricultural sector suggest that policy ideas in the Cuban economic model may have shifted in this manner. It remains to be seen whether these ideas have changed simultaneously in the three variables of policy and thus constitute a 3rd_{order change, or whether they have} changed incrementally.

As indicated in the research question, we want to explain the nature of the changes involving less state intervention, more economic liberties and the introduction of market based strategies in Cuba’s agricultural sector. One would expect certain exogenous factors – such as climate conditions and international market conditions – might have a major impact on the outputs of the Cuban agricultural sector. At the same time, it seems likely exogenous factors are difficult to anticipate, since they originate from outside the institutional environment.

It can be expected that changes in the Cuban economy are strongly related to the performance in the agricultural sector and overall food production. Thus, it

(23)

seems likely when exogenous factors affect the agricultural sector and the overall food production, it could also influence change of the policy paradigm. In the case of a vulnerable or ‘weak’ policy paradigm, exogenous factors seem likely to trigger a process of 3rd_{order change. Such a process might be} facilitated or accelerated by the presence of endogenous factors in the institutional environment – such as inefficiency in the Cuban economy – that pressure or undermine the policy course of the existing paradigm.

The following hypotheses are formulated in relation to Hall’s model:

a) If a weak policy paradigm is observed, then it is likely that exogenous factors trigger a process of 3rd_{order change.}

A ‘weak paradigm’ is characterized by the indicators that are likely to precede a paradigm shift, namely ‘positional advantages’, ‘debate on who holds authority over policy issues’ and ‘policy experiments or failures’.

b) Exogenous factors make a process of 3rd_{order change more likely, if the} existing policy paradigm is pressured by endogenous factors in the

institutional environment.

1.3.2 Model two – Mahoney & Thelen

Gradual Change Framework

Mahoney & Thelen (2010) have developed the so-called Gradual Change Framework (GCF), that explains how institutional change occurs on a gradual basis. In the GCF they aim to model continuous change, rather than (over)emphasizing the causal effects of specific moments in time. They argue that critical junctures – as exogenous sources of change – can be important to explain institutional outcomes, but they may also obscure why institutions are vulnerable to change in the first place (ibid.: 5). That is to say, institutions themselves can be the cause of institutional outcomes, even if these outcomes have been affected by critical junctures. It means that causal effects are not necessarily attributed to the exogenous environment of institutions; they may also be attributed to the endogenous environment of institutions.

(24)

Two assumptions are important for this argument. First, institutional outcomes are not necessarily the result of environmental – exogenous – events, nor do they have to be discontinuous of nature (ibid.: 7). This means that institutions can also change incrementally in a gradual process. Institutions may adapt to the changing institutional environment over time, without being replaced by new institutions or institutional rules. To illustrate: the NATO alliance between western states was originally intended to provide security against the common threat of the Soviet Union during the Cold War. However, despite the imminent collapse of the Soviet Union, NATO kept existing; its right of existence was redefined over the course of the 1990’s to include a broader mandate of military security, amongst others the stabilization of Eastern European (NATO, 2012). Second, institutional outcomes do not necessarily reflect specific interests of persons or groups. Conflict over institutional resources among actors with different objectives, may result in compromises that have unintended consequences (Mahoney & Thelen, 2010: 8). These conflicts are likely political of nature; their outcomes depend on actors’ relative position of power. To illustrate: a department of agriculture may have the power to invest in agricultural technologies, but the department of finance – managing public spending – may affect the actual decisions that are made. If reversed, this is probably less likely the case.

In the GCF, a feedback loop between power relations and the allocation of resources is assumed. These resources can be political or material means that are formalized into institutions, rules and procedures (ibid.). The loop works as follows: power relations between institutional agents determine how resources are distributed. Such ‘resource considerations’ inevitably lead to uneven allocation; certain resources will be allocated to specific (institutional) actors and purposes, while others will not. In turn, the allocation redefines existing power relations between institutions; the feedback loop is completed.

The GCF theorizes how these power relations take shape in institutions, by addressing which properties of their (political) environment – endogenously or exogenously – motivate different change strategies and lead to different types of institutional change. There are two elements that shape institutional outcomes; first, the characteristics of the political context can affect outcomes. This is reflected by the way the context allows actors to have veto possibilities within, or between institutions. These are abilities actors have to prevent institutional

(25)

change, or to direct such change. A high degree of veto possibilities means that it will be difficult to challenge or replace existing institutional rules, whereas few possibilities entails the opposite. Second, characteristics of institutions can affect institutional outcomes. This is reflected by the way institutional agents are allowed discretion to enforce rules and the freedom they have to interpret them. The level of ambiguity of a rule, and the interpretation of it, determine these factors (ibid.: 18). According to Mahoney & Thelen, the combination of veto possibilities and the level of discretion to interpret institutional rules will lead to different types of institutional change.8

The type institutional change layering has been selected to investigate in the case of Cuba. Layering means that new institutional rules are based on previous ones, involving small adjustments, but they do not necessarily replace the original rules or institutions (ibid.: 15). Such small adjustments may accumulate over time and eventually lead to bigger (policy) change (ibid.: 17). Displacement is the opposite; rules or institutions are completely replaced by others, similar to Hall’s 3rd_{order change. The Cuban state seems to develop new agricultural} strategies, while the institutional system appears to remain unchanged. This aspect is interesting to investigate, as it could indicate that a process of layering is occurring in the agricultural reforms of Cuba.

In the GCF so-called change agents are the catalysts for institutional change. They can be individuals such as politicians, scholars that are an expert in certain policy areas. Such (institutional) change agents are driven to change existing institutional structures. Rocco & Thurston (2014) assess that layering is likely to occur through change agents that already operate in the existing institutions. Change agents are capable to introduce small changes within the institutional context, but they are not able to change it as a whole. It means that the overall institutional system remains unchanged, while small changes or additions to existing institutional rules are made that lead to a gradual process of layering. The following section outlines three explanatory variables that are important to the process of layering.

Veto players and veto possibilities

Actors in the institutional and political system who hold authority to approve outcomes, or otherwise have the capacity to steer the policy process, are

(26)

defining for how institutional change through layering comes about. Tseblis (2011: 19) defines such actors – veto players – as follows: “Veto players are individual or collective actors whose agreement is necessary for a change of the status quo.” Institutional veto players are those individual or collective actors in the political system that derive their ‘veto possibilities’ from a country’s constitution (ibid.). They are granted powers to change, enforce, or implement institutional rules. Veto players have discretionary possibilities, or access to intellectual or material capacities allowing them to block institutional change, or drive it. These veto possibilities depend on the political or institutional context. It is important to note that institutional collective actors do not necessarily represent a single uniform veto player. For example, a parliament may be occupied by different parties. In such cases, the party holding the majority is the actual veto player, or what Tsebelis refers to as ‘partisan veto player’ (ibid.). In cases with many (collective) veto players in the institutional environment, it becomes increasingly difficult to change the status quo. This is because agreements need to be reached on a consensual basis, in spite of different preferences (ibid.: 39). Identifying which actors in the institutional setting are (actual) veto players in the policy process, is the first step in distinguishing between the kind of veto possibilities they have, and the distribution of possibilities among different (institutional) actors.

Institutional ambiguity or malleability

Institutional rules, or institutions themselves, may vary in terms of clarity and ambiguity. The level of ambiguity provides insight into how institutional outcomes vary. Following Rocco & Thurston (2014: 41), this is called ‘initial institutional ambiguity or malleability’. If an institution or rule is ambiguous it means that there are few institutional rules that constrain the actions of the institution. For example: a policy that states ‘the reduction of greenhouse gasses’ makes a wider variety of measures possible, than a policy that states ‘reduction of greenhouse gasses using renewable energy sources’.

Discretional freedom

Institutional outcomes may also vary because there are variations in the discretion institutional actors have to implement – or enforce – policies

(27)

according to their own interpretations (Mahoney & Thelen, 2010: 21). This variable is defined as ‘discretional freedom’ and overlaps with the previous indicators of ‘institutional ambiguity’ and ‘veto possibilities’. The level of ambiguity of a rule may determine how policy is effectively interpreted and carried out by institutional actors. If the rule is specific, it leaves little or no room for different interpretations, or the way it should be enforced. A high level of discretional freedom is the opposite, meaning that there is more room for variation in the interpretation or enforcement of the rule.

Important to note here, is that the discretion to enforce, or interpret, will also depend on the veto possibilities that are available to an actor. This means that differences in the interpretation and the enforcement of rules not only coincide with the level of ambiguity. They may also depend on the allocation of veto possibilities in the institutional environment and the access of actors to specific discretionary capabilities (to enforce and/or implement rules).

Layering in relation to the variables

With layering, the initial ambiguity (or malleability) of the institution or rules is expected to be low. This restricts change agents to replace existing rules or institutions, so that they can only introduce small adjustments (ibid.: 41-42). To do this, they must have discretionary capabilities, intellectual or material resources to “carry out modest changes at the margins of existing policies” (ibid.: 46) Material and intellectual resources may be used to lobby or convince veto players with discretionary capabilities that institutional change is necessary. It is expected that veto players are not able to prevent modest adjustments at these margins, as they do not involve significant changes of the status quo. At the same time, however, veto players may limit the discretional freedom of change agents through their veto possibilities. Conclusively, with layering it is expected that change agents have access to the discretionary capabilities of veto players, or are veto players themselves.

Hypotheses on gradual change

Endogenous and exogenous factors form the reasons for change. The institutional agents determine the ‘necessity’ and/or ‘need’ for change, while the nature of change is determined by their mutual power relations and the associated distribution of resources (as suggested by the feedback loop).

(28)

Mahoney & Thelen (ibid.: 2) emphasize the role of endogenous factors in relation to gradual change, since they may act over a relative long period of time. Gradual change is more likely to be related to endogenous factors than to exogenous factors, because endogenous factors act within the institutional environment and can be influenced and anticipated by institutional agents. Moreover, incremental changes can mitigate the effects that follow from endogenous factors. So, if in a process of change, the institutional agents and their mutual power relations remain unchanged by these endogenous factors, then this process will likely have a gradual nature.

If institutional agents and their mutual power relations are characterized by strong veto possibilities, a low level of initial ambiguity, and a low level of discretional freedom, then this gradual change seems likely to display the characteristics of layering. That is to say, new rules or institutions are placed atop or alongside existing ones, instead of replacing them.

Following the mechanisms and indicators that have been described under model two, the following hypotheses are formulated:

a) If a process of institutional change is observed, while institutional agents and their mutual power relations remain unchanged, then this process is likely a process of gradual change.

b) If strong veto possibilities, a low level of ambiguity, and a low level of discretional freedom are present in the institutional environment, then it is likely endogenous factors trigger a process of layering.

(29)

1.3.3 Comparing the models

Moving forward with both models: it seems logical to assume that ideas over policy may change over time. Hall’s policy paradigm is intended to address not so much institutions per se, but the way ideas behind the rules have changed. It is difficult to formulate concrete indicators that specify why these ideas change. They can only be derived from the results of actual policy (objectives) that define the overall policy paradigm, for example socialist policy objectives. Because this is more abstract, Hall’s typologies seem better suited to identify exogenous sources of policy change. Periods of sudden change are associated with 3rd_{order change and mostly explained as a result of exogenous factors or} shocks.

Contrary to the policy paradigm model, the GCF emphasizes gradual change by looking predominantly at endogenous factors. It makes the interaction between political context and institutions more concrete endogenously, by describing how change agents and veto possibilities among institutions affect one another. Nonetheless, the policy paradigm may affect the policymaking process endogenously – in institutions – as well. The political context of institutions is not only shaped by exogenous influences, but also – endogenously – by policy ideas and objectives. The policy process itself, can be examined from different viewpoints depending on the (institutional) context; namely, endogenously:

between institutions, or from the environment in which the institution functions.

Exogenously: from outside the institutional environment.

It is interesting to juxtapose layering to Hall’s 3rd_{order change since these}

institutional types allow us to assess policy from different perspectives, namely non-radical institutional change versus radical institutional change. Because these perspectives are mutually exclusive, the scope of this case study is

(30)

broader than if we were to look from only one institutional perspective. In the Cuban case, it is interesting to examine if agricultural policymaking is an example of either of these outcomes, or perhaps both. The aim is to identify which endogenous and exogenous factors play a role in whatever sort of institutional change is occurring in Cuba.

In this sense, the (historical) institutionalist models can complement each other and transcend the institutional labels they use to identify different kinds of institutional change. Specifically, why is it that Cuba has introduced several decentralizing and market-based measures in its agricultural sector? And how is it that this process seems to have taken place primarily in the last ten years? Both models have been visualized in Figure 1 and include a simplified representation of the mechanisms that have been outlined. Their respective indicators have been numbered and will be used throughout the following chapters. The arrows represent the direction in which influences act. For instance, the feedback loop of model two is represented by the two-way direction between ‘resources’ and ‘institutional agents’. In model one, social learning is represented by the arrow between ‘institutional agents’ and ‘policy’. Exogenous factors have been mentioned explicitly, while endogenous factors act within the institutional environment.

(31)

(32)

(33)

Chapter 2 Methodology and operationalization

The first section of this chapter presents the method of inquiry that is used in this case study, namely process tracing, and discusses which sources and data are used for the analysis. This is followed by the explanandum, wherein the nature of the agricultural policy change is outlined. The agricultural policy measures that have been taken are part of the explanandum, but will be discussed in detail in the next chapter. The explanandum is followed by the operationalization, which starts out with the selection of the institutional agents that have been selected to investigate the agricultural institutional change in Cuba, for what reasons and how they relate to Cuba’s institutional environment. Continuing, the concrete operationalizations for endogenous and exogenous factors are presented, after which the operationalization of the empirical indicators and institutional mechanisms – 3rd_{order change and layering – are} outlined for each model.

2.1 Research methodology and the use of sources

This thesis can be characterized as a single case study, or – more specifically – a single country study (Gerring, 2006; Landman, 2008). It means that the unit of analysis – change in Cuba – is used as the single example to investigate or make assumptions, about the theoretical framework that is used. As such, there are several methodological implications for how this study may be used and should be seen in a wider context.

Landman (2008: 86) points out that in comparative (political) research, single country studies involve intensive research on particularities that play a role in a specific country, while they are less extensive in their scope of analysis (since they concern only a single country). At the same time, intensive research of such particularities may be helpful to pose new questions in the research field, or expand on the applicability of certain theories in specific cases. So-called outliers or ‘deviant’ cases may lead to new theoretical questions. Thus, single

(34)

country studies may serve two comparative purposes: first, to provide information and contextual description on political developments or phenomena that may be useful to other cases. Second, they may be used to gain further understanding of (causal) mechanisms described in theories, through testing of specific outcomes, or by creating new typologies for (existing) conceptual designs (ibid.).

In this study both purposes apply; on the one hand, the transitional phase Cuba is in, would be difficult to understand without an understanding of how specific circumstances in the agricultural sector relate to Cuba’s political and economic context. Inevitably, this raises questions about the pieces of evidence that need to substantiate this aim. Empirically speaking, the decision-making process is only perceivable through the actual outcomes of policy. To illustrate: the Cuban

Lineamientos (guidelines) of 2011 are a result of some political process in Cuba.

The result of this political process may be regarded as a fact. However, when the political process is analyzed and theory is translated to practice, some degree of interpretation and bias from the researcher are always present. For this study, it means that the paradigm concept by Kuhn and Hall’s policy paradigm require more interpretation than Mahoney & Thelen’s GCF, as their concepts are based on thought structures and ideas. This makes it more difficult to research it empirically.

In terms of generalizability a single country study will have its limitations, since the specifics of one case – time, context and interpretation of evidence – are not one on one comparable to the other (Landman, 2008: 93). Even so, Alvesson & Skoldberg (2009: 21) assert that generalizing “beyond the empirical base” depends on your epistemological view, that is what one considers the basis for knowledge that is true. Continuing, if a view accepts “non-observables in the form of patterns and tendencies, common to and underlying several surface phenomena, successive expansions of the empirical area of application within a certain domain are both possible and desirable, even in qualitative studies” (ibid.: 22) In this study, it is accepted there are non-observables when it concerns the policy process and change. Thus, generalization on the basis of this case does not necessarily involve that its outcomes are representative of other cases. Rather, it follows from expanding the empirical base of the institutionalist models and their area of application in the policy process.

(35)

2.1.1 Process tracing

In the previous section the purposes of single country studies have been established. Also, it was stated that specific outcomes are likely not applicable to other cases. Therefore, it is important that the internal validity of a single country study is assured (Gerring, 2006: 43). It means that what is intended to be investigated in the case is actually analysed. Otherwise, it would defeat the purposes of the single country study. This holds especially for researching mechanisms described in theories, wherein causal relations between a given set of variables often play an important role. But through what kind of evidence do we establish the presence of these (causal) relations?

A technique common to (single) case studies is process tracing, which will be used in this thesis as well. Gerring characterizes process tracing as follows: “multiple types of evidence are employed for the verification of a single inference – bits and pieces of evidence that embody different units of analysis (they are each drawn from unique populations).” (ibid.: 173) Single country studies – such as this thesis – rely on contextual information and deductive reasoning to establish patterns or causal relations between variables (ibid.: 172).

For process tracing this means that different types of contextual information may be used as evidence to verify a certain pattern or causal relations, such as in hypotheses and theories. In the case of this thesis, this amounts to so-called ‘thick description’. This means that the contextual information will be used to provide highly detailed accounts of specific circumstances or situations that may have been relevant to the Cuban agricultural policy process. This increases external validity, as different types of contextual information can contribute to a better understanding of the phenomenon that is researched.

In practice, such information may include both quantitative and qualitative sources, such as documents from experts, legal documentation, policy documents, but also media coverage and historical accounts. Important to note is that these ‘bits and pieces of evidence’ are non-comparable to one another; their relative value cannot be determined for multiple samples – as is the case with quantitative research – because they are unique to the sample that is analyzed (such as Cuba). Thus, the value of evidence lies not in the amount of

Continuity and change in Cuban policy making during the 21st century