Understanding Public Policy Through the Lens of Political Regimes: The Case of Episodic Change in Agriculture Policy in Burkina Faso (1983–2022)

Abstract

This article addresses episodic change in agriculture policy in Burkina Faso over time. It aims to explain radical changes to this policy under the presidencies of T. Sankara (1983–1987) and B. Compaoré (1987–2014), as well as incremental changes to it under the administrations of M. Kafando (2014–2015) and R. Kaboré (2015–2022). The data come from a review of the literature on public policy and political regimes in Burkina Faso; grey literature; the texts of laws and press articles; and around thirty semi-structured interviews performed between September 2021 and September 2023 in Ouagadougou and Bobo-Dioulasso. Using the political regimes’ approach as a starting point, the article supports the argument that a regime's authoritarian nature, reflected in the state's failure to adapt to environmental pressures and in the society's weak policy capacity, plays a decisive role in radical policy change. By contrast, a regime's democratic nature, reflected in the state's efforts to adapt to environmental pressures and by the society's significant policy capacity, explains incremental change in the policy.

Introduction

As a branch of political science, the sociology of public policy has become more and more technical and distanced itself from macro-sociological aims (Bezes and Pierru, 2014: 494). But methodical and theoretical synergy with the other branches of political science offers a promising heuristic for these larger aims (Darbon and Provini, 2018). This article contributes to the process of broadening the study of the sociology of public policy by analysing changes in agricultural policy under several political regimes. The literature on change in public policy draws on several analytical perspectives to study change. Hall (1993) conceptualises change as occurring at three levels or orders, which are associated with different types of change. Change in public policy can be radical when it alters the hierarchy of goals behind policy (third-order change). Change is incremental when it is affected through the adoption of new policy instruments (second-order change) or adjustments in how policies are implemented (first-order change). Third-order change leads to radical shifts, while the two other orders of change induce marginal adjustments.

Another approach to studying change is to focus on four interdependent dimensions: ideas, instruments, institutions, and interests (or actors) (Boussaguet, 2020: 42). Institutionalist approaches emphasise the power of institutions to impose restrictions, formal or informal, on actors’ room to manoeuvre and on possible present and future policy alternatives (Lindblom, 1959; Pierson, 2000). Idea-based approaches, meanwhile, emphasise the transformative role of systems of representation and interpretation, which guide actors in their choices when change begins (Hall, 1993; Muller, 2013; Sabatier, 1988). Actors’ approaches stress the actors’ potential to cause gradual, transformative changes in a restrictive institutional environment (Mahoney and Thelen, 2010).

Scholars examine the links between political regimes and public policies from many angles. The neo-institutionalist approach shows that constitutional arrangements (parliamentary and presidential regimes and so on) determine the strategies and resources of public-policy actors (Knoepfel et al., 2016: 99–120). The concept of public-policy styles is another angle from which we can understand the links between political regimes and public policy. Among other things, this concept deals with the use (or not) of consensus, as well as the degree of interest groups’ involvement in public policy (Richardson, 2014). Furthermore, it is useful to emphasise the contribution of work done on instruments (Halpern et al., 2014). There is a correlation between political regime and instrument types (communication-based, incentive-based, regulation-based, etc.). Citizen participation, the effects of political changeovers, and electoral cycles are additional research subjects that directly or indirectly establish links between political regimes and public policy (Couture and Jacob, 2019; Weaver, 1986). Public policies have also been studied in the contexts of neo-patrimonial regimes (Bach and Gazibo, 2011) and transitional regimes, particularly in the African context (Saidou, 2020).

This article adds to this literature on the links between political regimes and public policy by focusing on the processes of change. According to some authors, incremental changes are the norm in liberal democracies, while inertia and radical changes are more likely in hybrid regimes (Jones et al., 2019; Sebők et al., 2022). This article uses this insight to consider changes in agricultural policy in Burkina Faso. Since the 1980s, Burkina Faso's agricultural policy has been punctuated by the episodic return of agroecology to the institutional agenda. ¹ Agroecology is defined simultaneously as a scientific discipline, a collection of agricultural practices, and a social movement, all sharing an alternative concept of agriculture to the industrial version (Wezel et al., 2009). It is guided by principles of social justice, a non-compensatory view of the environment, food sovereignty, and empowerment of farmers.

In Burkina Faso, Capt. Thomas Sankara's 1983–1987 regime, called the National Council of the Revolution (CNR), directed its agricultural policy towards agroecology via reforms that promoted vulnerable populations’ access to farm inputs, the agricultural market, and public services (Zahonogo et al., 2023). This shift led to the replacement of the traditional chieftainship in its role of allocating rural land by new actors: the Committees for the Defence of the Revolution (CDRs). The CNR used instruments of communication to mobilise citizens to complete public work projects (large and small water retention facilities). It also used binding instruments to raise purchase prices for producers, nationalise land, and prohibit importing food and agricultural products (Tallet, 1996). The radical change introduced by the Sankara regime in the sector contributed to increasing tensions between the regime and civil society actors, including farmers, unions, and journalists, which led to the regime's overthrow reversal in October 1987 by Capt. Blaise Compaoré (Kyélem de Tembèla, 2012).

Following the country's agreement in 1991 to a Structural Adjustment Program (SAP), the Compaoré regime (1987–2014) reoriented agricultural policy towards using liberalisation to achieve food security. Regime legislation prioritised agribusiness as its agricultural model of choice for the country (Lavigne Delville and Thieba, 2015) and genetically modified (GM) cotton as the preferred type of cotton to be cultivated in Burkina Faso beginning in 2008 (Batenburg, 2020). (Hereafter referred to as GM cotton or Bt cotton, the specific variety grown in Burkina.) The priority given to cash crops also led to the use of incentive tariffs to encourage food imports, through the adoption of World Trade Organisation rules and West African Economic and Monetary Union agricultural policy (Fouilleux and Balié, 2009).

Following Compaoré's fall from power in 2014, agricultural policy has undergone changes of the first and second orders only. Under the transition government led by Michel Kafando (2014–2015) and the presidency of Roch Kaboré (2015–2022), new legislation, ² institutions, and actors ³ were included in a liberalised agricultural policy that recognised agroecology as one of the means by which food security would be achieved (Sib et al., 2023). GM cotton was abandoned in favour of conventional cotton paired with sensible use of chemical inputs and sustainable land management. This reform process was similar to the dynamic seen in Senegalese agricultural policy (Milhorance et al., 2023).

In sum, the dynamic of agricultural policy in Burkina Faso over time is characterised by a radical shift to agroecology under the Sankara regime (1983–1987), a radical reversal away from agroecology under the Compaoré regime (1987–2014), and an incremental return to promoting agroecology under the Kafando (2014–2015) and Kaboré (2015–2022) regimes. This series of changes in agricultural policy is at the heart of the present analysis.

On the theoretical level, the task is to explain the production-related changes in agricultural policy through the lens of the political regimes. More concretely, it is to understand these regimes through the concepts of the state's adaptative capacity and the society's policy capacity ⁴ (Boda and Patkós, 2018; Jones et al., 2019). Thus, the four political regimes are analysed comparatively, using these two variables. The state's adaptative capacity refers to the “government's ongoing responsiveness to the citizens’ preferences” (Collura, 2019: 321). It evaluates the government's ability to avoid “very large and erratic changes” in public policy by adapting to environmental pressures using incremental changes (Jones et al., 2019: 9). Societal policy capacity refers to the extra-governmental mechanisms and actors that create and disseminate key information and knowledge on policy issues to policy makers (Boda and Patkós, 2018: 15–16). It accounts for the participation of civil society in the process of creating and implementing public policies.

Using Gazibo's (2010) proposed criteria for classifying regimes – that is, the degree of competition and the style of leadership – as a starting point, the literature identifies the Sankara regime (1983–1987) as a populist regime ⁵ (Gazibo, 2010) and that of Compaoré (1987–2014) as semi-authoritarian ⁶ (Hilgers and Loada, 2013; Hilgers and Mazzocchetti, 2010). The Kafando regime (2014–2015) is classified as a democratic transition government, because of its political contests’ competitive nature (Natielsé, 2020; Saidou and Bertrand, 2022). ⁷ The analysis is long-term, covering agricultural policy over a period of about 40 years (1983–2022). The theoretical decision led to the formation of the central question: how much is the degree of change in agricultural policy shaped by the state's adaptative capacity and by society's policy capacity?

Methodologically, the empirical material has been constructed from two data sources. The first is an analysis of the expert literature on the public policies and political regimes of Burkina Faso, the grey literature, the texts of laws, and press articles. The second is information collected from around 30 semi-structured interviews conducted between September 2021 and September 2023 in Ouagadougou and Bobo-Dioulasso. These interviews focused on researchers, parliamentarians, leaders of political parties, civil-society organisations, farmers’ organisations, and the private media. The sectors of agricultural policy under analysis were, essentially, the governance of the seed sector, the fertiliser and pesticide sector, and the land sector. Structured in two main parts, the article examines agricultural policy through a double lens, analysing the state's adaptative capacity in the first case and society's policy capacity in the second case.

Agricultural Policy Through the Lens of the State's Adaptative Capacity

Radical changes in Burkina Faso's agricultural policy occurred, we contend, when the state was unable or unwilling to respond or adapt to citizen preferences. In both cases, the nature of the regime influenced this responsiveness, which we refer to as adaptative capacity. Under the Sankara and Compaoré regimes, very little compromise occurred with interest groups, consequently, policy was developed unilaterally, leading to more radical policy changes. Unilaterally developing public policy. In contrast, the nature of politics under the regimes of Kafando and Kaboré resulted in greater attention to the interests of a variety of actors, which produced incremental change in agricultural policy.

Radical Change Caused by the State's Failure to Adapt

Radical changes in Burkina's agricultural policy occurred under two regimes with different ideological viewpoints: socialist for Sankara and liberal for Compaoré. In both cases, their lack of openness with regard to civil society allowed them to radically reform agricultural policy. Their lack of responsiveness to interest groups’ demands, i.e. the state's low adaptative capacity, meant that competing interests were excluded from the institutional agenda, which ultimately limited the beneficial results of these reforms.

We can see the lack of consultation in these regimes’ unfavourable reactions to the demands of actors who disagreed with the regimes’ decisions on agriculture. For Sankara, the objectives of agricultural policy included not only food security for the most vulnerable, but also the liberation of the farmers from traditional and economic elite domination. Traditional leaders attempted to preserve their control over land. Their objective was to maintain control over farmers and more generally to preserve the traditional land tenure system. By contrast, the Sankara regime considered this traditional system as a form of injustice against farmers. By nationalising the land, the Sankara regime meant to democratise access to land and prevent traditional leaders from using land rights to exploit farmers. The major economic actors claimed that imports made an essential contribution to food security, but the Sankara regime was convinced that these actors were more concerned with profit than with meeting the nutritional needs of vulnerable households (Labazée, 1988; Poussart-Vanier, 2005), so the regime opted to ban the import of agricultural and food products. Similarly, rather than take the advice of international lenders and financial institutions, like the World Bank, Sankara, an advocate of endogenous development, rejected the Structural Adjustment Programmes (PAS), through which the international financial institutions promoted economic liberalisation. His government also rejected these entities’ financial aid, which he considered an instrument of political domination over developing countries (Ziegler and Rapp, 1986).

As for Compaoré and his political allies, they considered the family agriculture model promoted by farmers groups as an archaic system to be avoided and as a serious obstacle to the achievement of food security. ⁸ According to Salif Diallo, Minister of Agriculture from 2000 to 2008 under Campaoré and agribusiness entrepreneur in the American and European model, family agriculture “is pessimism. These family operations that we are talking about today, all together, produce no more than two or three European or American farmers do. The farmers’ community is all well and good, but it needs another dimension, agricultural entrepreneurism, to move forward.” ⁹ His position was representative of the government's clear policy preference for agribusiness actors in the National Policy for Rural Land Tenure Security in Burkina Faso (PNSFMR), which codified the links among land tenure security, agribusiness, and food security in ways that benefited these dominant actors (Thieba, 2010). That the government offered only symbolic distribution of inputs to vulnerable farmers ¹⁰ sent a clear message that the government would not make family agriculture a part of its policy on agricultural development. Instead, it chose to emphasise the cotton sector, agricultural entrepreneurism, and genetically modified crops. Mobilisation by civil society groups against the decision to cultivate genetically modified (GM) cotton in Burkina Faso failed. ¹¹

Faced with the Campaoré government's radical shift, several African citizens’ movements praised Sankara's former CNR for its agricultural policy based on family farming, which had enabled the country to achieve food self-sufficiency in 4 years (FAO, 1992). Nonetheless, it would be inaccurate to label the radical change under Sankara a complete success. Under Sankara, agricultural policy basically consisted of government initiative. The lack of citizen involvement in setting policy meant that the regime adopted inconsistent measures. Notably, these included increasing purchase prices paid to producers while simultaneously lowering pay for government workers, which plunged city dwellers into chronic insolvency. The government's effort to monopolise food sales was also difficult to implement. The Sankara regime had trouble collecting and paying for crops in the villages and could not prevent sellers from exporting food. In addition, the process of allocating rural land, in which the CDRs were involved, was paralysed because of the tacit yet powerful opposition of traditional leaders who had been excluded from the process (Tallet, 1996). Citizens’ criticisms of government policy and process were blocked by leaders in the CDRs, ¹² who were military personnel loyal to the CNR. Finally, the principle of democratic centralism – meant to “encourage communication and the expression of opinions from the base towards the peak (that is, from the CDRs towards the CNR) and vice versa” (Martin, 1990: 50) – appeared “to function in only one direction: from the peak towards the base” (Martin, 1990: 56).

Decision-making was similarly centralised under the Compaoré regime. According to certain actors, the decision to introduce genetically modified (GM) cotton came directly from the president. According to them, this decision was based less on economic motives than political ones. ¹³ To support their argument, these actors pointed out that GM cotton was introduced before the adoption of legislative and institutional instruments needed to determine whether the decision was appropriate (COPAGEN, 2017), and they also cited the president's deep involvement in setting the agricultural agenda (Traoré et al., 2014). Interest groups – in particular farmers’ organisations through their umbrella group the Farmers Confederation of Faso (CPF) – deplored the ineffectiveness ¹⁴ of the National Farmers Day (JNP) initiative, ¹⁵ which they decried as an insufficient government response to their pleas. ¹⁶ Their critiques stemmed from unkept promises and the JNP's co-optation by administrative and political actors. ¹⁷ The lacklustre government response to farmer interest groups implies that the regime did not consider dealing with the farmers’ widespread dissatisfaction to be crucial to the regime's own survival. Although elections were held under the Compaoré regime, farmers voters were subject to manipulation because of illiteracy, the parties’ failure to provide political education, and traditional chiefs use of charismatic domination to sway the votes of farmers (Vaast, 2010). In this context, the punitive purpose of the vote was poorly understood by farmers, who constituted the majority of the population and the electorate. Compaoré's re-elections and his regime's ability to withstand the various crises that affected society, including the farming community, in 2008 ¹⁸ and 2011 (Hilgers and Loada, 2013) seem to support the hypothesis that voting did not make farmers an important factor in Campaoré's survival in office. By contrast, following the fall of the Compaoré regime, the political issue of development assistance and the public's assent to the governmental agenda would act as incentives for the Kafando and Kaboré regimes to adapt to environmental pressures.

Incremental Change as a Reflection of the Political Regimes’ Openness

The incremental changes in agricultural policy under the Kafando and Kaboré regimes were expressed in the adoption of new instruments (LOASP, PNSR2, SND-AE ¹⁹ ) and new actors. For example, a National Correspondent for Agroecology (Correspondant National de l’agroécologie) was designated within the agricultural ministry as part of the Agroecology Program of the Economic Community of West African States (ECOWAS). In the implementation of this programme in member countries, the correspondent acted as the direct liaison with the program's management unit, one of whose expected outcomes was the integration of agroecology into national and regional policies. The post-Compaoré context was one of expanding democratic spaces, in which the government was more attentive to social demands coming from actors who had diverging opinions and interests regarding agricultural policy. Political leaders efforts to seeking political gain, along with the pursuit of international aid, guided the peripheral inclusion in the institutional agenda of new national and international perspectives that favoured the agroecology model. The Kafando ²⁰ and Kaboré ²¹ regimes’ recognition of agroecology, as well as their commitment to make it an instrument for achieving food security, may, however, have been primarily strategies to attract international funding. ²² Certain defenders of agroecology perceive these regimes’ decision to promote agroecology not as a true commitment but rather as a pragmatic promise they knew would be possible to fulfil given that technical and financial constraints effectively mean that almost all family farming in Burkina Faso is based on agroecological practices. ²³

The recognition under the Agro-Sylvo-Pastoral Production Act (LOASP; adopted during the Kafando regime) ²⁴ of divergent agricultural models and the composite definition given to agroecology in the National Strategy for the Development of Agroecology (SND-AE; adopted during the Kaboré regime) ²⁵ show that decision-makers were concerned with reconciling conflicting interests. ²⁶ These interests belonged to the coalition of agroecology actors ²⁷ on the one hand, and on the other hand, the coalition of conventional-agriculture actors, ²⁸ who still enjoy considerable influence. The influence of conventional-agriculture actors leads agroecology defenders to argue that the national strategy's definition of agroecology is an excuse to continue using conventional approaches to agriculture. According to one of these actors,

what is reasonable ²⁹ for me might not be reasonable for you. […] Take an authorisation to utilise chemical inputs […]. Right now, I could say that 100 kg per hectare is sensible for me, and someone else will say that 200 kg is sensible for his land. ³⁰

With regard to land, although the Kafando and Kaboré regimes recognised rural land management as a major public problem (CRNR, 2015), they were under pressure from the beneficiaries of the agribusiness policy, and they did not include the question of rural land in their institutional agenda. Under Compaoré, these leaders themselves had been entrepreneurs in, ³¹ and beneficiaries of, ³² agribusiness. During the Campaoré years, the recognition of private rural land, which the state confirmed by delivering property titles such as certificates of ownership of rural land facilitated the arrival of “new actors,” or “agrobusiness men,” in the agricultural sector (Hochet, 2014). The practice of informal land acquisition by the “new actors” preceded Law No. 034-2009/AN on rural land tenure, which provided for these titles. (For example, in the Hauts-Bassins and Centre-Ouest regions, the practice dated back to the first decade of the 2000s.) As a result, most of the agrobusinessmen were able to formalise their land rights with the law because of their de facto possession of the land. Meanwhile, many common-law landowners – concerned about generating income or fearful of their land falling into the hands of local collectives – yielded to the siren's song of land commodification. In this commodification, the high bidders were the agrobusinessmen, whose considerable financial capital distinguished them. But although land tenure security (which was promoted by the land law of 2009 and its instruments) seems to have had positive effects on agricultural investment and production, its impact on food insecurity remains unclear. The marketing and production decisions of agrobusinessmen do not necessarily coincide with vulnerable populations’ purchasing power and food priorities (GRAF, 2011; Sanfo et al., 2024).

With regard to seed systems, although the Kafando and Kaboré regimes upheld the suspension of GM cotton cultivation as requested by the agroecology movements, they appear nonetheless to have made concessions to agricultural biotechnology actors by authorising trials of other genetically modified seeds, such as cowpeas)(or niébé. As confirmed by a member of the Institute for Economic and Social Development (INADES): “Yes, we withdrew [genetically modified seeds] for cotton, but we have now introduced them for niébé. […] We worked on it at INERA [the Institute for the Environment and Agricultural Research], whose facility was very well protected; surely they have started distributing these seeds by now.” ³³ This experiment is highly valued by actors in biotechnology, and the resulting seeds’ distribution is, indeed, imminent. ³⁴ Furthermore, legislation on biotechnology – which does not ban the import of genetically modified products – was not revised during Kafando and Kaboré's presidencies, either. In sum, policy shifts under the two presidents were mixed. Although allowing more room for agroecology, policy support for conventional and agrobusiness agricultural practices persisted. This mixed outcome reflected, we argue below, the capacity of societal actors to shape policy under the more competitive regimes of the two presidents.

Agricultural Policy as Seen from the Perspective of Society's Policy Capacity

The Sankara and Compaoré regimes’ failure to adapt was correlated with the limited policy capacity of the society. Societal actors could not manage to counterbalance the decision-making power of political authorities. By contrast, the Kafando and Kaboré regimes’ openness to political pluralism strengthened interest groups linked to agroecology movements, which allowed them to bring agroecology back into the institutional agenda, even as existing agrobusiness actors and conventional farmers used their societal position to shape the agenda at the same time.

Powerless Interest Groups Faced with Radical Change

Under the Sankara and Compaoré regimes, society's policy capacity was limited in part because farmers’ organisations in Burkina Faso were founded as civil-society organisations whose legitimate (even legal) scope of action was limited to development measures, ³⁵ not political action (Barry, 2006). Association movements, such as the federation of Naam organisations, the SOS Sahel non-governmental organisation (NGO), and the Inades-Formation, which were led by agents of the Sakara administration ^{36 ,} contributed to the establishment of the CNR's agricultural and environmental policy, ³⁷ consequently they were unlikely to oppose its agenda. In addition, bilateral aid, which was the main source of financing for the regime's development policy, went through NGOs focused on development, including farmers’ mutual-aid organisations. For example, Sankara entrusted French ecologist Pierre Rabhi with developing a national policy based on agroecology (Roger, 2015) because this agricultural model presented an opportunity to achieve the regime's objective of food self-sufficiency. Rabhi, in turn, oversaw the Centre for International Relations among Farmers for Development (CRIAD), which trained the leaders of Burkina Faso's first farmers’ agroecology organisations to spread the model around the country (Roger, 2015). Under the Compaoré regime, local actors in development continued to spread agroecology, both by helping new farmers’ organisations emerge and by helping build networks among these organisations (Sajaloli et al., 2013).

While the Sankara administration directly supported the development of agroecologically-oriented farmers’ associations, it limited opponents’ ability to mobilise against their agricultural policy. Under the Sankara regime, the traditional chieftaincy lost its political and administrative privileges, and the major agricultural traders lost an important source of revenue – that is, imports of certain agricultural and food products.

Sankara's CNR worked to limit opposition influence via parties, unions, and the media. Opposition parties were suppressed and members arrested, ³⁸ while left-wing political movements – already close to Sankara – were represented in the government (Martin, 1990). Unions were weakened as well with the creation of Committees for the Defence of the Revolution (CDRs) in workplaces; a Leninist strategy to transform independent unions into docile instruments of the regime (Phelan, 2016: 64). Although the right to strike was not abolished, the creation of the CDRs within workplaces and the targeted investigations of union leaders and factions deterred would-be strikers. As, civil servant unions were weakened, farmers – now organised by the regime – were called upon to replace them as the new societal. Finally, after the closure of the private daily paper L’Observateur following a fire, suspected to be the work of the CNR (Ziegler and Rapp, 1986), the regime also controlled the nation's media (Bianchini and Koala, 2003).

Economic interest groups began using religious spaces as a forum to subtly express their opposition to the CNR's austere policies (Labazée, 1988). Meanwhile, the traditional chieftaincy placed its trusted agents in the CDRs in an attempt to control land management and used its traditional rites to delegitimise the CDRs in the villages (Lalsaga, 2020). Nevertheless, these efforts to exert influence did not dissuade the Sankara regime, which – although aware of their existence – remained faithful to and all-in on its revolutionary ideals.

While the Sankara administration tamed agroecologically oriented activism and farmers’ organisation through incorporation and organisation, while repressing opponents, the Compaoré government sought closer collaboration with groups advocated genetically modified production and agribusiness while repressing any opposition to its radical shift in policy away from agroecology. The capacity of movements defending agroecology, which included mainly advocacy movements, researchers, and political parties committed to agroecology, environmental protection, and human health was limited. were. Their political action began only in the 1990s, as an indirect consequence of structural adjustment programmes (SAPs). Farmers’ organisations were not formally included in evaluating the impact of SAPs on agriculture, leading to impact assessments and conclusions that contradicted farmers’ lived experiences. Consequently, farmers decided to establish networks to gain recognition as legitimate actors and stakeholders in agricultural policy at national, regional, and international levels. (McSween, 2013; Mercoiret, 2006).

Such is the case with the National Agro-Pastoral Workers Union (SYNTAP). This advocacy movement was formed in 1998 in the country's west to denounce the government's lax approach to the financial cost of fighting white-fly infestations, which the Burkina Faso Society for Fibres and Textiles (SOFITEX) burdened cotton producers with in 1997–1998. ³⁹ According to its founder, SYNTAP is an independent representative of cotton producers, whose official representative – the National Union of Cotton Producers (UNPCB) – seemed to back cotton companies and the government, particularly on the matter of introducing GM cotton in Burkina Faso. ⁴⁰ In response to the approval process for GM cotton, which was begun in 1993 ⁴¹ and completed in 2008, farmers’ movements, ⁴² researchers, ⁴³ and political parties ⁴⁴ expressed their opposition to the decision via messaging about the possible health and environmental consequences, and via threats to sue and calls for the collective rejection of the decision (COPAGEN, 2017; Ouédraogo, 2006).

This opposition was not tolerated by the government, which used informal means to limit these actors’ influence on the decision-making process (Batenburg, 2020; Luna and Dowd-Uribe, 2020). According to Issouf Traoré, a member of the Coalition for the Protection of Genetic Heritage (COPAGEN), certain leaders in the cotton sector incited people to boycott these actors’ efforts to raise awareness of GMOs. Some of the researchers who opposed the “hasty” introduction of GMOs were forcibly transferred to workplaces far from Ouagadougou. ⁴⁵ For example, Ousmane Tiendrébéogo blames Minister Salifou Diallo for the difficulties SYNTAP faced in obtaining official recognition from the state from 1997 to 1999. In particular, he says that the minister prohibited press organs from publicising SYNTAP's actions and kept economic actors from financing its activities. ⁴⁶ The agroecology movements’ political actions may not have prevented the introduction of Bt cotton, but they nevertheless forced the government to respect the preconditions provided for in the documentation on the matter. Notably, these included the creation of a national agency, an observatory, and scientific committees on biosafety; the adoption of legislation on the issue; and the conduct of experimental studies stipulated in the Cartagena Protocol, which the government signed in 2003 (COPAGEN, 2017). Overall, the agroecology movement did not have the space or capacity under the Campoaré government to fundamentally challenge its radical policy shift, which allowed for a radical change unlike that seen under the more competitive and open regime that would follow his ouster from office.

Incremental Change Driven by Agroecological Movements

Agroecology movements seized the political opportunity that followed the fall of Compaoré to actively encourage the return of agroeconomy to the political agenda. These movements are examples of what Braud (2023: 48–49) calls support groups, because they exist to argue for a specific topic. ⁴⁷ To achieve the suspension of GM cotton cultivation, the agroecology movements – judging that the political context was favourable – opted to participate in global anti-GMO demonstrations by organising a march in Ouagadougou on 23 May 2015. In advance of the march, a memorandum with supporting evidence ⁴⁸ was delivered to the authorities. More than a thousand people gathered for this march (Naveau, 2018). Because of its size, the march drew the public's attention, forcing decision-makers to adopt a meaningful response. Without the political support that it had had under the Compaoré regime, and backed into a corner by the anti-GMO demonstrations, those in charge of the GM cotton issue ⁴⁹ saw that they had to suspend GM cotton cultivation and return to conventional cotton.

To successfully push for the legal recognition of agroecology, the agroecology movements had to use other resources in addition to demonstrations and scientific data. Specifically, they worked to take over relevant decision-making arenas (Pralle, 2003) as a means of shifting the political agenda and aligning it with prevailing views on the issue of genetic modification, which had changed thanks to the anti-GMO march and the suspension of Bt cotton cultivation. For example, the selection process for members of the National Transition Council (CNT) ⁵⁰ used candidate interviews, which allowed one agroecology actor to reach parliament, Ousmane Tiendrébéogo, a founding member of the Citizens Collective for Agroecology (CCAE). He became an MP on behalf of the farming community thanks to SYNTAP, of which he was the founding president. To force the CNT to respond to Mr Tiendrébéogo's request that agroecology be included in the Agro-Sylvo-Pastoral Production Act, the CCAE settled on the strategy of participating in the global marches, with the help of its partners outside the country. Ousmane Tiendrébéogo shared the details of this event: “I had problems at the CNT. During oral questions, when I would submit my questions, they were never chosen, and my term was almost over. So I created the event [the march against Monsanto]. Thankfully, our partners from France were there.” ⁵¹

It was also thanks to a group of about a dozen parliamentarians, whom researchers equipped with information on the importance of farmers’ seed systems, that the parliament assented in 2017 to the agroecology movement's request that it adopt a law to protect the country's genetic resources for plant breeding. ⁵² The decision to create a national agroecology strategy document was driven by the national correspondent of the ECOWAS agroecology program, ⁵³ but the credibility enjoyed by some agroecology and sustainable-agriculture entrepreneurs gave them authority in the writing of the document and its action plan. ⁵⁴ Agroecology entrepreneurs also became more and more involved in political parties, which helped strengthen the parties’ policy capacity. The partnership between the CCAE and the Balai Citoyen (Citizen's Broom), which manifested in the Balai Citoyen's participation in the 2015 anti-GMO march, ⁵⁵ was extended via a connection with the Servir et non se servir (To Serve Others, Not Ourselves) movement, a political entity founded by Guy Hervé Kam, a leader of Balai Citoyen. ⁵⁶

The entrepreneur Salif Ouédraogo's membership in President Kaboré's party ⁵⁷ doubtlessly facilitated his elevation to the head of the Ministry of Agriculture. His rise to the top of this ministry specifically opened politics to the agroecological agenda and generated support for the new national agricultural strategy (SND-AE). Salif Ouédraogo had previously been a development engineer and operations director at SOS Sahel. ⁵⁸ As noted earlier, this NGO has long been a proponent of agroecology. It, led by Alfred Sawadogo, a long-time agroecology entrepreneur who was director of the NGO Monitoring Bureau (BSONG) under the Sankara regime. He was also one of the actors who conceived of the village water-retention approach, thanks to which the Sankara regime was able to achieve food self-sufficiency (Verdejo, 2020). Beginning with the popular uprising of 2015, Sankara's ideas had momentum thanks to civil-society actors such as the Balai Citoyen. Conscious of the prominent place Sankara occupies in the national and international collective imaginations, agroecology entrepreneurs have made sure to frame their advocacy as a call to return to the CNR's agricultural model. ⁵⁹ The Kafando and Kaboré regimes, in their quest for legitimacy, used the same logic. To win popular support, they attempted to link themselves to the Sankara legacy. So, the Kafando regime, for example, revisited pending court cases, including the one on the Sankara assassination. The Kaboré regime, for its part, re-adopted some CNR policies, such as the plan to use phosphate deposits for agriculture. Certain leaders of Sankarist parties justified joining the president's political circle by referencing President Kaboré's appropriation of certain agricultural plans from the Sankara regime. For example, Bénéwendé Sankara declared that President Kaboré was the Sankarist standard-bearer. ⁶⁰ The combination of new organising on the part of agroecology proponents and their entry into politics, paired with Burkina Faso's new leaders’ need for popular support and desire to link themselves to Sankara's legacy helped bring agroecology back into the political agenda, even as other societal and economic actors ensured the pendulum did not swing in the direction of radical reform.

Conclusion

This article has shown the heuristic value of focusing on political regimes to explain changes in public policy. By considering the state's adaptative capacity and the society's policy capacity under different regime types, we can better understand why radical or more moderate public policy change occurs.. The case of Burkina Faso agricultural policy over time shows that the degree of change is intrinsically linked to the nature of the political regime. Specifically, under each regime, we should assess the state's adaptative capacity and society's policy capacity. The radical change under the Sankara and Compaoré regimes, driven by new ideas about agriculture, came about in the context of a double failure: the state's insufficient ability to adapt to pressures from social movements and these movements’ inability to influence the governing powers. Thus, this result indicates that radical change, which affects public-policy goals, is more likely in regimes that are to some degree authoritarian, where relations between the state and interest groups are characterised by conflict.

Marginal changes, which affect public-policy instruments, occur in more democratic societal configurations, whose regimes are more susceptible to the influence of interest groups’ wishes. This is what we retain from the stories of the Kafando and Kaboré regimes’ influence on agricultural policy. The incremental changes under these regimes were thanks to the efforts of agroecology entrepreneurs, even if the opening of the political currents also benefited competitors, helping to limit the scope of the changes that took place. This shows that it is not enough to open a “window of opportunity,” as Kingdon says (2003), for radical change to occur. The competition between interest groups with relatively equal capacities tempers the perspectives of paradigmatic change. This result calls us to expand Kingdon's model by focusing on the effects of competition between interest groups to influence the institutional agenda on a specific issue.

This article revives the debate on interest groups’ interactions with the state (Offerlé, 1998). It corroborates the argument put forth by Grossman and Saurugger (2006), according to which public-policy change relies on the configuration of the relationships between interest groups and states. These authors distinguish between three “ideal,” “typical” categories: the statist, pluralist, and neo-corporatist models. State-society relations under the Kafando and Kaboré regimes were, to some degree, of a pluralist configuration. Nevertheless, the case of Burkina Faso is distinct in some regards from the model developed by Grossman and Saurugger. The Sankara and Compaoré regimes may have resembled the statist model, but mobilisation was nearly absent from the interest groups’ repertoire of tactics. Clearly, the study of changes in agricultural policy strengthens the relevance of an approach that lies at the intersection of public policy analysis, interest group sociology, and the sociology of political regimes. The synergy between different research fields within political science, with a detour into the macrosociological aspects of political life, remains fertile ground for understanding the measures states take to regulate modern societies undergoing change.