Decentralisation or Recentralisation in Bolivia? Autonomous Territorial Entities and Intergovernmental Relations in a Decentralised State

Political history in contemporary Bolivia is profoundly marked by regional cleavage. If the 1952 Revolution brought to the fore the profound class contradictions, which meant the emergence of new political subjects (working class and popular sectors), in the construction of a new political order, the National-Popular State, characterised by a strong nationalism, but also by centralism that pursued social equalisation; the 1960s witnessed the emergence of strong regional mobilisations that confronted state centralism¹. This gave rise to a demand for regional autonomy that would be one of the constants of the most recent Bolivian politics and embodied in the decentralisation process that took shape in the 1990s. The following pages attempt to explain this process, focusing mainly on the period after the 2009 Constitution.

The political and administrative decentralisation process, together with the development of strong mechanisms for citizen participation and its parallel democratisation at local government levels, generated a solid territorial complexity and density and explained the advance toward a highly decentralised state in the 1990s (Del Campo, 2007). The approval of the new Constitution in 2009 and the territorial distribution of power changes it brought deepened the decentralisation process. However, the reality has been different. The autonomy of sub-national governments and administrations has been strongly conditioned by (a) the political hegemony achieved by the ruling party, the Movimiento al Socialismo (MAS), which has influenced the emergence of local leadership and their postulation at the national level; (b) normative centralism, which is embodied primarily in the legislative development after the Political Constitution of the State and the Ley Marco de Autonomías y Descentralización “Andrés Ibáñez” (Autonomy and Decentralisation Law - LMAD); (c) fiscal inequality that has established significant differences between large capitals and rural municipalities, as well as an under-financing of the meso levels of government; and (d) a malfunctioning model of intergovernmental relations, which has not been able to cohere the complex Bolivian sub-national system.

The representation crisis began taking shape in Bolivia in the late 1980s when the country’s founding party of the 1952 Movimiento Nacionalista Revolucionario (Nationalist Revolutionary Movement - MNR) implemented its first structural reforms. The closure of mining companies, the privatisation of public companies and natural resources, and incentives for the internationalisation of agricultural production were seen by a large part of the population as a deviation from the majority’s interests. The continuation of reform policies by all governments in the 1990s, both on the traditional left [Movimiento de Izquierda Revolucionaria - MIR (Movement of the Revolutionary Left) - and MNR] and on the right [Acción Democrática Nacionalista (National Democratic Action) and], meant that the protests that were initially aimed at the MNR were eventually directed at the party system as a whole. Thus, the system of pacted democracy, the result of the indirect second round of the presidential election, which had tolerated continuous corrupt practices, began to be understood by the population as a perverse mechanism that served the elites of the major parties to negotiate their interests and quotas (Grindle and Domingo 2003). Thus, voters increasingly perceived that parties had become mere intermediaries of political patronage.

Thus, decentralisation, the recognition of original procedures or the introduction of new instruments and figures of control favoured the population’s participation; however, they did not silence the demands for change. On the contrary, the socialisation of excluded groups into logics of power through the new participatory mechanisms, and the prolonged inability of institutions to offer attractive economic and political alternatives, meant that demands that initially called for specific changes were transformed into demands for the redefinition of the institutional system as a whole.

The approval in 1994 of the LPP opened up the Bolivian political system on two fronts: firstly, it reduced the barriers that hindered the entry of small, geographically concentrated political parties by creating a new level of political competition, the municipal level. On the other hand, the LPP signified the beginning of a new political trajectory from local to national leadership (the case of Manfred Reyes Villa in Cochabamba).

Similarly, the LPP allowed for municipal participation in the resources of tax co-participation. However, this eminently municipal process² left the pattern of political-administrative relations at the intermediate level incomplete. The law redefined municipalities’ boundaries to cover rural and urban areas (which allowed the municipalities’ jurisdiction to be extended to the whole territory), recognising the legal personality of indigenous communities and peoples, peasant communities and neighbourhood councils as Grassroots Territorial Organisations (OTBs).

Although demands for greater territorial autonomy had been essential and recurrent in Bolivian political history (primarily through the regional Civic Movements or the indigenous marches), at that time, decentralisation was a public policy designed and implemented from above. Moreover, this movement presented a clear political and ideological orientation, which sought to unblock the functioning of the state (throwing the ball down to the lower level of government³) in a context of institutional reforms anchored to the neoliberal economic model. In this sense, Barja et al., (2012) point out that this top-down model led to the creation of a transfer system strongly oriented towards solving horizontal fiscal inequalities rather than promoting inter-jurisdictional competition.

However, decentralisation took a different course than initially envisaged. The deepening of this process reinforced the national representative institutions’ crisis by increasing citizens’ expectations of state institutions dominated by traditional political parties. Therefore, if the process of deepening political-administrative decentralisation generated democracy, it was also one of the instruments that accelerated its “political decay”.

Since the 2005 elections and especially since the approval of the new Constitution in 2009, it has been pointed out (Zegada and Brockman, 2016) that there has been a process of strengthening the meso levels of government, which sought to weaken the municipal political weight and the growing protagonism and departmental leadership, especially in the East. The political role they have been playing, together with the emergence of other levels of political and administrative autonomy, redefined the role of local governments.

The Political Constitution of the Bolivian State (CPE) established four political-administrative levels at the territorial level: departments, regions, municipalities and indigenous territories (TIOs) (Table 1). For the former, the Constitution foresaw two alternative scenarios: autonomy or administrative decentralisation, leaving the decision to the popular will expressed through a referendum. In the case of the TIOs and municipalities, on the other hand, autonomous management could only be carried out through autonomous governments. The situation of the provinces and regions was left to later legislative development, especially the latter, whose definition, by mandate of Article 280 of the Constitution, depended on the LMAD.

Bolivia is thus constitutionally configured as a Plurinational Unitary State model with 347 autonomous territorial entities (ATEs), including nine autonomous departmental governments, 326 autonomous municipal governments, 11 municipalities in conversion to Indigenous Native Peasant Autonomy (AIOC) and one regional autonomy. This number can increase since the constitutional norm permits new territorial units administered by new ATEs⁴.

The Bolivian autonomy model differentiates between four types of competencies (CPE Chap. 8, Article 297): privative, exclusive, concurrent and shared. Additionally, there are nine catalogues of competencies for the different levels of government (Articles 298–304).

Privative competencies are those where the legislation, regulation and execution are not transferred or delegated and are reserved for the Central level of the state (Article 297/I/1).

Exclusive competencies are those in which one level of government has legislative, regulatory and executive powers over a given subject matter and can transfer regulation and implementation (Article 297/I/2).

In the case of the exclusive competencies assigned to each of the ATEs, the allocation is rigid, i.e., symmetrical or homogeneous; this implies that each level of government must exercise, both mandatorily and gradually, the competencies assigned to it by the Constitution, but taking into account the institutional capacities of that governmental entity. However, although the allocation of powers is symmetrical, the truth is that the territorial, institutional, organisational, economic and financial reality among the different autonomous governments is asymmetrical (Ortuño Cassón, 2020:187). In many cases, their institutional capacity is feeble and heterogeneous, making it difficult to achieve the expected results in public management and frustrating the expectations created among the citizenry.

Concurrent competencies are those where the complete legislation corresponds to the Central level of the state, and the other levels simultaneously exercise regulatory and executive powers (Article 297/I/3).

Furthermore, shared competencies are those subject to basic legislation of the Plurinational Legislative Assembly, whose development legislation corresponds to the ATEs, according to their characteristics and nature. Regulation and execution will correspond to the ATEs (Article 297/I/4).

It should be noted that nothing was determined about the supervisory power, while the definition of Basic Law was left up in the air, leaving the limits for this type of regulation to the free interpretation of the Plurinational Legislative Assembly.

On 19 July 2010, Law No. 030 Marco de Autonomías y Descentralización “Andrés Ibáñez” (LMAD) was enacted to regulate the state’s autonomy regime and define the guidelines for the actions of sub-national governments, consequently delimiting their capacities, rights and responsibilities. The Law’s scope was enormous, as it sought to lay the foundations for the territorial organisation of the state, the different types of autonomy, the procedures for access to autonomy and the elaboration of Statutes and Organic Charters, the competency systems, economic and financial matters, the mechanisms and instruments of coordination between the Central level of the state and the ATEs, and define a general framework for participation and social control in the latter (Article 3, LMAD)⁵. This law allowed the departmental governments to design their autonomous statutes to outline their institutional framework, which is different from the national level and other departments, in response to the heterogeneous political, economic and social reality of the different territorial entities.

However, the LMAD adopted a centralist stance (Ortuste, 2016), endorsing harmonisation and uniformity in developing autonomous processes⁶. The Plurinational Legislative Assembly attributed itself Constituent Assembly status, going beyond the provisions of Article 271 of the CPE. In many cases, it came to regulate the exclusive competencies of the autonomous governments “to the point of breaking down the subject matter on which they were based and imposing limits on their actions”. Additionally, concurrent competencies distorted and modified the constitutionally accepted definition so that the central level could assume regulatory and executive powers jointly with the ATEs, “creating a parallelism and duplication of functions” (Herrera Acuña, 2021: 120–121).

In the shared competencies, the LMAD referred to fundamental laws, which would later allow, through government majorities in the Plurinational Legislative Assembly, to take advantage of the legislative power of the national executive and invade the regulatory and executive powers of the ATEs (Ortuste, 2016; Herrera Acuña, 2021).

Thus, of the four types of competencies defined in Article 297 of the CPE, the central level has direct interference in the exercise of legislative powers in three types of competencies, which means that the ATEs have reduced decision-making spaces in the definition and implementation of public policies in their territories (Zegada and Brockman, 2016).

The legislative development of the LMAD has led some authors to speak of normative centralism (Urenda, 2017). Thus, assessing the LMAD itself, Urenda points out that this law regulates the competencies of the central level and does not establish any autonomisation of competencies process. “More than a law of autonomies, it is a law of centralised control of the autonomous process” (Urenda: 2017:72). In the same direction, legislation such as the Lottery and Gambling Law No. 60 (25 November 2010), a shared competence, but which nevertheless reserves to the state the regulation, supervision, collection and allocation of lottery taxes; the Avelino Siñani-Elizardo Pérez Education Law No. 70 (20 December 2010), which re-centralises education, giving the departmental governments only the concurrent competence of education management, and leaving the construction of school infrastructures to the municipalities; the Law on Classification and Definition of Taxes and Regulation for the Creation and/or Modification of Taxes of Autonomous Governments No. 154 (14 July 2011), which curtails the tax creation powers of ATEs; the General Law on Telecommunications, Information Technologies and Communication No. 164 (8 August 2011); the Law on the National System of Citizen Security No. 264 (31 July 2012), which treats the competence of citizen security as exclusive to the central level of government; the Law on the Promotion of Investment in Hydrocarbon Exploration and Exploitation No. 767 (11 December 2015); and the Law on the Comprehensive State Planning System and its Subsystems No. 777 (21 January 2016), which establishes centralised State Planning (Urenda, 2017: 67–86).

In many processes of political decentralisation, the deepening of decentralisation leads to the emergence of sub-national political parties, which sometimes have no interest in reaching national positions. Some parties attempt to influence local politics or pressure national government parties to achieve specific objectives. In other scenarios, decentralised deployment allows for the emergence of party subsystems with their arenas, delimited by their spheres of competence in most cases. In other cases, the hegemony of one or several parties at the national level means that the nationalisation of local politics is very important, and the spaces for local political autonomy are very narrow.

In the Bolivian case, there are multi-level national parties with differential degrees of presence and territorial gravitas, such as MAS and, to a much lesser extent, Unidad Nacional (UN), and, on the other hand, parties of sub-national territorial origin that constitute organisations of resistance to the governmental party. Although in this case, they are temporary and volatile coalitions in terms of acronyms and composition, which vary in each electoral process (Zegada and Brockman, 2016:56) and especially in each territory (department/municipality).

If we analyse the sub-national elections that have been held since the arrival of Evo Morales to the Bolivian presidency in 2005, we can say that until the 2004 municipal elections, the presence of traditional political parties continued to be necessary⁷. As O’Neill (2005) pointed out, electoral interests have motivated different decentralisation strategies from the centre. Indeed, there has been progressive electoral bleeding of the MNR in national elections in Bolivia since the mid-1990s⁸. This decline was also noticeable at the sub-national level.

In the 2004 municipal elections (see Table 2), MAS won 112 municipal mayor’s offices (34.2%)⁹, MIR 29 (8.8%), MNR 25 (7.6%)¹⁰, ADN 22 (6.7%), UN 19 (5.8%), MSM and MIP 12 each (3%) and PP 6 (1.8%). It is true, however, that the change in the electoral system that allowed citizen groups to run meant that out of a total of 327 municipalities, 254 (77.6%) were won by political parties, 54 (16.5%) by citizen groups and 19 (5.8%) by indigenous peoples¹¹. The latter only achieved electoral success in the departments of La Paz, Oruro and Potosí, as in the rest of the country, their presence was relative. Meanwhile, citizens’ groups showed greater strength in the departments of Pando and Tarija. It should be noted that traditional parties such as the MIR and MNR still had a certain presence at the municipal level. However, it is clear that these elections allowed for the possibility of significant growth of the MAS in the 2005 presidential elections and show “a municipal electoral competition highly autonomous from the national” (Ascarrunz, 2021a:20).

Concurrently, the December 2005 general elections coincided with the popular election of prefects for the first time in Bolivian history, from nine constituencies by a simple majority. As shown in Table 3, the positions were relatively evenly distributed in these elections. MAS won the prefectures of Chuquisaca, Oruro and Potosí, while PODEMOS (the opposition alliance) won those of Beni, La Paz and Pando. The prefectures of Cochabamba, Santa Cruz and Tarija presented their own departmental proposals.

In this complex electoral context, in July 2006 (and simultaneously with the elections for constituents), a referendum on departmental autonomies was called to define the characteristics of political decentralisation. In both cases, MAS won with an absolute majority, although in four departments, the autonomist option won (Beni, Pando, Santa Cruz and Tarija).

Although MAS won resoundingly in the 2010 municipal elections¹², it did not obtain the same votes as at the national level, with an average drop of 14 points (Deheza, 2012:41). This result was mainly due to the multiplicity of local forces built around municipal leadership and the growing discontent of union leaders who were not qualified as MAS candidates. MAS was defeated in Santa Cruz, Beni and Tarija, invariably opposition regions, and in seven of the nine largest cities: La Paz, Oruro, Potosí, Santa Cruz, Sucre, Tarija, and Trinidad.

While none of the opposition political forces had a national presence, some maintained some regional and local leadership, such as the MSM, although it too proved volatile over time. Moreover, since there were few landslide victories, most winners were forced to negotiate with the second forces in departmental assemblies or municipal councils or face a problem of ungovernability.

Sub-national elections for departmental (governors, members of departmental legislative assemblies) and municipal (mayors and councillors) authorities took place on 29 March 2015. All departments elected Governors, eight of them by simple majority. Only Santa Cruz established an absolute majority formula and a second electoral round.

In these elections, MAS lost the stronghold of El Alto to Unidad Nacional. Also, the governorship of the department of La Paz, went to the opposition, where Morales’ former dissident education minister, Felix Patzi, won 52% of the votes.

However, MAS won six governorships, leaving three for opposition forces. The election result also seemed to confirm that MAS had managed to penetrate the eastern lowland electorate: after the first and second round (which took place on 3 May of that year), MAS won the governorships of Pando and Beni, managed to force a second round in Tarija, and won 31% of the votes in the department of Santa Cruz.

MAS’s growing influence was not only related to the fragmentation of the political opposition and its difficulty in achieving a cohesive leadership but also to the fact that the Morales government nationalised the party’s reach in a favourable economic context, where sustained economic growth¹³ and an increase in public works managed to appease the traditional economic elites of the East (Sánchez, 2015). It has been noted that after 2009 and until the 2021 municipal elections, there was “a municipal electoral behaviour more in line with national trends” (Ascarrunz, 2021a:20).

A particular feature of the sub-national elections is the critical presence of ethnic-cultural pluralism, which is produced through the incorporation, across all political forces, of leaderships with an indigenous native peasant matrix (Zegada and Brockman, 2016:59).

On 7 March 2021, the election of the authorities of 336 municipalities, the only regional autonomy, and nine departments took place (four governorships went to a second round on 11 April 2021). In addition to these posts, the election of the authorities of seven indigenous autonomies was also defined. While MAS won 240 municipal mayor’s offices at the municipal level, it only triumphed in three departments. However, it reduced its vote compared to 2015, indicating that these results respond more to the fragmentation and drift of the opposition than to the strength of this political organisation (Ascarrunz, 2021b). Once again, a strong personalist political culture favoured political personalities¹⁴ beyond party acronyms.

Some authors, such as Morgan (2018), underline the negative impact of the decentralisation process on political parties and their decline in the face of emerging territorially based personalist leadership.

It would therefore seem that the results have moderated, albeit partially, the MAS’s hegemony. However, if we analyse it from a broader time horizon, we can say that the emergence of this party has essentially reconfigured the structure of competition in the party system, influencing and conditioning the rest of the opposition political options, both at the national and sub-national level¹⁵.

The results of the 2021 sub-national elections suggest that municipal political competition is gaining some autonomy from the national level. This autonomy may have to do with the changes that have taken place at the national level in party leadership, the emergence of political and social leadership with a greater degree of autonomy, but also, as indicated (Eaton, 2013), with a scenario of recentralisation of power in Bolivia. Thus, MAS hegemony seems to coexist with a high degree of fragmentation of opposition political forces at the municipal level, while departmental politics is more stable, with opposition strongholds being reduced to the departments of Santa Cruz and Tarija.

Only three of the nine departments have approved their Statutes of Autonomy: Tarija, Pando and Santa Cruz. Chuquisaca, Cochabamba, La Paz, Oruro and Potosí submitted their statutes for a referendum but were rejected by the population. The department of Beni’s statute is currently under constitutional review.

In the case of regional autonomy, in 2009, the municipalities of Yacuiba, Carapari and Villamontes in the department of Tarija acceded to regional autonomy, finally becoming the autonomous regional government of the Gran Chaco in 2016.

Of the 339 municipalities, only 15 organic charters have been approved in the referendum, and ten have been rejected. Finally, in the case of Indigenous Native Peasant Autonomy, Charagua in Santa Cruz (2016) and Uru Chipaya (2017) in Oruro have managed to become an Indigenous Native Peasant Autonomous Government via municipal conversion. Of the 212 rural municipalities, 11 are at some stage of the autonomy process. Only Raqaypampa has been constituted as a GAIOC out of the more than 300 existing territories of Indigenous Native Peasants. Notably, only three indigenous autonomous regions have completed the process.

Since ATEs are not subordinate to each other and have equal constitutional status (CPE, Article 276), all levels of government are obliged to interact and dialogue (both vertically and horizontally) to improve the management of public policies throughout the territory.

Thus, the LMAD (Title VII) contemplates the establishment of six intergovernmental coordination mechanisms: the National Council of Autonomies for political coordination (CNA); the State Autonomies Service (SEA), in charge of technical coordination and strengthening autonomy management; the State’s Integral Planning System, an instrument for programmatic, economic and social coordination; the Sectoral Coordination Councils (CCS), technical norms and instruments for financial coordination and intergovernmental agreements and conventions between the different ATEs. Of these six, only CAN and CCS require multilateral negotiation (Rocabado, 2017). Similarly, the institutional dismantling of the Ministry of Autonomies itself from 2017 onwards should be highlighted, subordinating the ministerial portfolio to a Vice-Ministry dependent on the Ministry of the Presidency, a politicised body per se.

None of these processes has taken place smoothly and permanently over time. It seems clear that the objectives of MAS in the national government do not coincide with the interests of regional political forces. In the first case, the interests of the party-state prevail, while territorial issues seem to prevail in the second.

As has been pointed out previously (Ibero-American Report 2021 of RIGS), there is a lack of systematic institutionalisation of intergovernmental cooperation bodies, both at the constitutional and especially the legal level. In many areas, these bodies are replaced by political initiatives dependent on ad hoc bodies and on the greater or lesser political leadership of those who convene them at the federal or regional level (Rocabado, 2017).

Regarding financing the autonomy process, once the Constitution was approved, changes to the fiscal decentralisation regime were made at the legislative (laws) and executive (decrees) levels, with very little coordination with the sub-national levels.

Articles 104 to 107 of the LMAD establish the resources of departmental, municipal, indigenous native peasant and regional ATEs. Although each autonomous government has different sources of resources, transfers from the National General Treasury and royalties represented 60.4% of total resources in 2020, mainly composed of tributary Co-participation, direct taxes on hydrocarbons (IDH) and royalties (Ortuño Cassón, 2020:103). These resources significantly benefit the most populated departments and the regional participation criterion in exploiting natural resources. The municipal governments are the ones that have obtained the most extraordinary funding, with 63.5%, due to the number of municipalities, the crucial resources they receive (tributary co-participation, IDH) and greater revenue-raising powers. The resources administered by the capital municipalities plus El Alto accounted for around 50% of the total resources reported in the period under analysis. In the La Paz and Santa Cruz departments, more than 50% of the resources are concentrated in the capital cities (Ortuño Cassón, 2020:110).

This observation indicates that the large municipalities enjoy the greatest fiscal autonomy, with a high percentage of their resources, and not the small rural municipalities, AIOCs or departments, which have practically no tax domain. Heterogeneity remains as many Decentralised Autonomous Governments are still reluctant to change local or departmental taxation because of the political-electoral consequences. Fiscal laziness and dependence on resources from the central government make fiscal decentralisation a goal rather than a reality.

This article’s main objective is to show how the Bolivian decentralisation process has halted in recent years, specifically since the approval of the LMAD. It can even be argued that the country has undergone a recentralisation. As has been pointed out for Latin America as a whole, this recentralisation has been driven by two elements: the ideological positioning of the state’s central role and the emphasis on the fiscal costs of decentralisation (Cravacuore, 2015).

In the Bolivian case, the explanation is multidimensional. First, it comes from the hegemonic political weight the ruling MAS party attained, which has dramatically conditioned the emergence of local leadership and their postulation at the national level. Second, the development of normative centralism is mainly reflected in the legislative development following the LMAD, which recentralises several of the competencies shared between the central level and the ATEs. Third, the unequal access to funding by the ATEs has led to large differences between large capital cities and rural municipalities, underfunding of the meso levels of government, and finally, the malfunctioning of the model of intergovernmental relations. Consequently, progress has been slow and with few results.

Finally, the political interests of the ruling party have subordinated the decentralisation process, resulting in a politicisation of the system as a whole.