Party favours: How political parties hobbled decentralisation in Uruguay

Decentralisation has become a sine qua non of political legitimacy across the developing world. Buoyed by promises of administrative efficiency, meaningful citizen participation and decreased central expenditure, decentralisation has supporters of nearly every political stripe.

During the populist wave from the 1930s to the late 1950s, Latin American states were highly centralized as civil-society groups were co-opted into national state machinery. Centralised government was maintained in the military regimes that followed. In the 1980s, the debt crisis and the military regimes' political failures rendered centralized states fiscally and politically unviable.

Decentralisation was the natural response to the foregoing. Yet its efficacy in improving the lives of the dispossessed varied across countries. The participatory budgeting of Porto Alegre, Brazil is lauded for its longevity and meaningful citizen participation (Veltmeyer, 2007). Argentina's fiscal decentralisation led to a substantial narrowing of health and education disparities between poor provinces and rich provinces (Habibi et al., 2003). Decentralisation in Ecuador, however, has been ineffective because it has been "characterized by a high degree of incoherence." (Faust, von der Goltz & Schloms, 2009, p. 41) Montevideo, Uruguay's decentralisation has had a mixed record - service-delivery gains are tempered by the persistence of low political participation among the poor.

I will answer the question "How effective is decentralization in improving the lives of the dispossessed?" by examining the role of political parties in shaping decentralisation. Using national- and municipal-level evidence from Uruguay, I argue that compromise and competition between and within political parties makes decentralisation less effective in improving the lives of the dispossessed.

Literature Review

Decentralisation is the "dispers[al of] fiscal, political and administrative responsibilities across different tiers of government and between the public and private sectors." (Litvack, Junaid & Bird, 1998, p. 26) Rondinelli (1980) differentiates between three degrees of decentralisation: deconcentration; delegation; devolution. Deconcentration is a mere transfer of responsibility, unaccompanied by authority or funds, from the centre to a local body. Delegation gives both responsibility and management authority over certain tasks to bodies governed indirectly by the centre. Devolution gives responsibility, authority and funding to local authorities, making them fully accountable to their constituents (Canel, 2001).

Decentralisation's development efficacy has been given rich theoretical and empirical treatment. McConnell (1966) wrote that decentralisation favours local elites because existing power relations condition how new power is distributed. This observation is borne out by the case of Ecuador, where governments of localities with greater income inequality were less likely to spend transferred monies on projects preferred by the poor (Araujo et al., 2008). Local political parties exploit inequality by engaging in what Herzer and Pirez (1991) identify as the shifting of decentralisation resources away from institutions and into clientelistic relationships. Popular dissatisfaction is multiplied when local governments lack the fiscal capacity to meet the inevitable upsurge in demands (Huntington, 1981).

Veltmeyer (2007) brings a Marxist perspective to the decentralisation debate, criticising it and the associated notion of popular participation as neoliberal ruses to avert the overthrow of capitalism by the masses. He is right to argue that decentralisation does little to address poverty and inequality's structural causes. While Uruguay’s economic collapse in the 1960s resulted from world market fluctuations, it does not follow that Uruguayans seek to end capitalism. It can be assumed that they prefer present-term improvements over the precariousness of a transition to a new economic system.

Despite Veltmeyer’s objections, decentralisation promises greater citizen participation and satisfaction. Oates (1972) argued that decentralisation increases citizen satisfaction by allowing local governments to adjust services to meet local consumption preferences, and strengthens citizen political involvement via local financing. In order to effectively represent citizens, local governments need the resources and authority to engage citizens outside of elections because voters’ choices in local elections often reflect their allegiances in national elections and not their local preferences (Dillinger, 1994).

Decentralisation, especially at the national level, is often a tool that parties use to consolidate their power. In 1991, Chilé’s governing leftist Concertación sought to democratise municipal government, where former dictator Augusto Pinochet had not institutionally entrenched the right. The rightists assented on condition that regional elections remain untouched (Eaton, 2004). Peru's president Alan García enacted decentralisation policies strictly to increase his party's representation in regional government (Kim, 1992). These cases suggest that “the most potent factor in the implementation of decentralization in developing countries is ‘political’ in nature and operation.” (ibid., p. 250)

In multiparty systems, political parties hinder meaningful popular participation in two ways: political parties inherently aggregate citizens’ preferences; they are motivated by short-term electoral gain and not by citizens' long-term interests. Regarding the former, Seabright (1996) writes that political elites’ preferred central-local power balance is not necessarily the balance preferred by the citizens whom they represent. The latter is exacerbated by interparty competition. The participatory budgeting of the Workers' Party (PT) in Porto Alegre, Brazil succeeded in part because the PT was the city's only credible political party (Goldfrank, 2007). The lack of competitors allowed the PT to orient decentralisation towards citizen needs without sacrificing electoral goals. Uruguay’s opposition parties used its leverage in the city government to ‘centralise’ the institutional setup of decentralisation put in place by the FA, its common rival. In cases where parties ultimately control the distribution of public goods, the marginal benefit a citizen gets from participation is low. That local governments’ budgets are, as a rule, not sufficient to satisfy citizen demands further discourages participation (Canel, 2010).

My essay addresses both the effect of politicisation on decentralisation’s development efficacy and on its capacity to engender popular participation by showing that the institutional setup of decentralisation can be highly influenced by interparty dynamics. Money and power were transferred from the national government to departmental (provincial) governments in a series of ad hoc compromises, for which there were no citizen accountability, made by the two major parties in efforts to stem the FA's rise. In Montevideo, direct intervention by the opposition parties resulted in institutional changes that severely compromised the ability of Montevideans outside political parties to influence government decisions. These two cases will be used to generate a theory about the relationship between political diversity and the efficacy of decentralisation for the dispossessed.

Evidence and Analysis

This section will discuss national-level decentralisation and Montevidean municipal-level decentralisation in Uruguay. For both levels, interparty compromises will be shown to have created particular forms of decentralisation that had limited effectiveness in improving the lives of Uruguayans in general and of the Uruguayan dispossessed in particular. At the national level, a reliance on central-departmental negotiation rather than on robust needs-based formulas in allocating downward transfers left Uruguay’s fiscal decentralisation little able to improve the lives of the dispossessed. Montevideo’s 1989 participatory budgeting programme was hampered by a 1993 compromise that entrenched the opposition in neighbourhood institutions and left citizens disillusioned with decentralisation.

National-level decentralisation

Uruguay is a highly centralised state. However, the centre has weak oversight over its subnational units. The Partido Colorado (Colorados) and the Partido Blanco (Blancos), Uruguay’s two ‘traditional parties’, have been able to exploit the accountability void left by the centre through their strong multilevel (national, departmental and municipal) organisations. The FA, their main challenger, has challenged them by developing robust machinery and constituencies at all levels. Uruguay’s unified election system, in which regional- and national-assembly elections were administered on the same ballot and in which voters could not split their ballot between parties (Eaton, 2004), made buttressing departmental party units a key election strategy. These electoral considerations were implicit in the 1990 and 1995 central-departmental fiscal transfers and in the 1996 constitutional reform.

FISCAL DECENTRALISATION

Despite strong de jure centralisation, the centre’s fiscal control over the departments was weak. Most departmentos accrued substantial deficits throughout the 1980 and 1990s.
By the end of 1989, the 19 departmentos owed a total of $7,604,000 USD. Most of that debt had been incurred with the national social-security system Banco de Previsión Social (BPS), which from 1985 to 1989 had financed over 20 per cent of the expenditures of 11 departmentos.

It was clear that the situation was not sustainable. In 1990, the recently-elected Blanco majority government arranged a departmental debt-forgiveness program in which the value of each departmento’s forgiven debt would be directly related to its share of reductions in nationwide departmental public employment. The money for these transfers was taken from the BPS. The Blancos sought to consolidate their constituencies in politically volatile departmentos, and, as a party with a majority national government, were able to mobilise the fiscal resources to do so.

By 1995, 14 departmentos had re-expanded their employment, some by over 30 per cent. The problem had grown too big for the Colorado minority government to ignore. Using monies provided by public utilities and the BPS, the government gave transfers proportional to departmental debt. The transfers effectively rewarded fiscal indiscipline. They also won the Colorados the cooperation of the Blancos (1), who proved useful allies in the 1994-1996 constitutional negotiations.

Although there is no data linking the BPS bailouts to development outcomes, it is likely that they impacted negatively on Uruguay’s poor. Firstly, subnational debt was reduced by impoverishing social security, a public good that covers 60 to 80 per cent of Uruguay's workforce (van Ginneken, 2003). The 1995 bailout was likely especially injurious to the masses because Uruguay's gross domestic product contracted 2 per cent in that year (Finch, 1998). Secondly, because national tax money paid for the allocations of higher-debt departmentos, residents of fiscally-disciplined departmentos subsidised the debt of other departmentos, which did not curb spending after the bailouts.

POLITICAL DECENTRALISATION

Uruguay's highly centralised state and party structures would seem to militate against decentralisation. Yet the results of the 1994 national election introduced a game-changing variable: the strength of the FA. Its 30.6 per cent of the popular vote sprung the Colorados (32.3%) and Blancos (31.2%), who together formed a majority (63.5%), into action. The Blancos agreed to vote for a run-off in the presidential election (which the Colorado candidate was expected to win) in exchange for a decoupling of subnational from national elections and a constitutional amendment calling on the government to formulate decentralisation policies, both designed to boost their performance in departmental elections. A Colorado presidency and Blanco departmentos would keep the FA at the municipal level, where it was hoped that it would fade into irrelevance. The splitting of elections and the nod to decentralisation gave subnational leaders grounds on which to demand more devolution from the centre. (2)

A limited decentralisation proceeded insofar as a highly centralised state and a lack of legislative clarity permitted it to do so. The constitutional reform created COSEDE, an intergovernmental advisory (not binding) panel on decentralisation policies. COSEDE negotiated with the government in 2000 a law reserving 3.13 per cent of national revenue for departmentos. The vagueness of Uruguay's 1934 municipal code, which regulates departmentos' provision of municipal services, has given departmentos the scope to spend transfer funds on health care, housing and socioeconomic development, which are legally under the national government's purview (Eaton, 2004).

Political decentralisation in Uruguay was more an admission of statutory imprecision and interparty negotiation than a turn towards citizens' demands. There are no national-level data linking the reforms to improvements in development outcomes, although Eaton (2004) notes that the reforms allowed the leader of the San José departmento to create jobs by working directly with foreign companies. It can be preliminarily concluded that political decentralisation and subsequent fiscal decentralisation led to improvements in services. However, the Blancos' lack of a pro-poor orientation makes their dominance of departmentos an obstacle to improving the lives of the dispossessed.

Municipal-level decentralisation

Most, if not all, of the literature on municipal-level decentralisation in Uruguay discusses Montevideo, the country's capital city, because it is the first city in which the explicitly pro-decentralisation FA took power. The FA effected its devolutionary decentralisation plan within a year of being elected to a strong majority in 1989. Once the municipal Colorados and Blancos had regrouped in 1993, the decentralisation institutions that had served the poor materially and politically were bureaucratised such that citizen participation became irrelevant to the operation of government. The post-1993 institutions gave the opposition institutional means of sabotaging the FA's reputation and future electoral prospects.

PRE-1990 DISPOSSESSION IN MONTEVIDEO

Following Uruguay's 1960s economic crisis, poverty and inequality in Montevideo rose sharply. The city centre was depopulated as people moved to outlying informal settlements throughout the 1970s (Goldfrank, 2007). In-migrants faced inadequate public services and social exclusion (Canel, 2010).

Economic crisis cast into doubt the Colorados' and Blancos' effectiveness, creating a constituency for a leftist solution. The emergence of any such solutions was delayed by the 1973-1984 military dictatorship. The 1985-1989 Colorado government was unable to bring broad improvements to the mass of Montevideans, 52 per cent of whom in 1988 rated the government as "bad" (Perelli, 1991, cited in Goldfrank, 2002). Its ratings had dropped despite a network of political clubs that would co-opt neighbourhood associations and mete out personal favours and public services (Goldfrank, 2007).

The FA, a leftist coalition including Socialists, Christian Democrats and even Colorado and Blanco defectors, capitalised on a latent leftism and on the Colorados' failures to take power in the 1989 election, inheriting a city with 27 per cent of households below the poverty line (Goldfrank, 2007). The leftism and neighbourhood organisations that had coalesced during the military dictatorship were seen as the keys to improving the lives of Montevideans (Winn & Ferro-Clerico, 1997).

DECENTRALISATION IN 1990: DEVOLUTION

In 1990, the FA unveiled the Centros Comunales Zonales (CCZs), which would serve as spaces for citizen deliberation and as service providers in each of the city's 18 zones (Canel, 2001). Each CCZ had a team of professional staff, processed routine administrative requests and held thematic commissions to collect project proposals for submission to the mayor's office (Goldfrank, 2002). The commissions, in which 60 per cent of neighbourhood associations participated from 1990 to 1993 (ibid.), thereby provided citizens with a modicum of direct political power.

The CCZs were effective in improving the lives of the dispossessed in terms of participation and service delivery. They made government more accessible by installing contact points in each zone, expedited the resolution of problems by having professional staff in each zone and implemented locally relevant projects. For example, the outlying zone of Colón built roads to its prime farming areas in order to give its farmers easier access to Montevideo markets (Winn & Ferro-Clerico, 1997). There were city-wide improvements as well: the number of illegal dumps dropped from 1,700 to 150 between 1990 and 1994; from 1985 to 1989, an average of 430,000 m2 of concrete was poured yearly to fix roads, and from 1990 to 1994 an average of 755,345 m2 was poured annually (Goldfrank, 2002).

The FA was able to carry out its devolutionary decentralisation because of an abundance of political will and popular support. The Colorado government's failures thoroughly discredited the traditional parties, which could not mount significant opposition to devolution of power away from party-controlled institutions. However, by 1993 forces in support of deconcentration had found their footing. The FA's socialist and communist wings were in revolt because they had expected the CCZs to organise FA mass fronts (Winn & Ferro-Clerico, 1997). The Colorado Party had strengthened its popular support and control of local politics by reviving the neighbourhood political clubs (Goldfrank, 2007). In response to opposition obstructionist tactics at city hall, FA mayor Tabaré Vázquez set up a multiparty commission to negotiate a mutually agreeable decentralisation (Canel, 2001).

DECENTRALISATION IN 1993: DECONCENTRATION

What resulted was a complex web of social and political institutions that recentralised decision-making power in each zone. On the social side were the Consejos Vecinales (Neighbourhood Councils), directly-elected advisory bodies each representing a barrio (neighbourhood) in each of the 18 zones. On the political side were Juntas Locales (Local Boards), which were five-member panels, with three government seats and two opposition seats, that had the final say over project proposal shortlists submitted to city hall. The mayor chose Junta members from lists submitted by parties (Canel, 2001).

As a result of bureaucratisation and politicisation, popular political participation dropped off precipitously. Even half of elected councillors did not show up for Consejo meetings (Canel, 2010). Because the Juntas had the final say on budget decisions, Consejos needed good relationships with Junta secretaries to bring government resources to citizen demands. These relationships sometimes existed - the 2000-2005 secretary for Peñarol barrio's Junta personally interceded to help residents turn vacant land into a public park (ibid.). Yet La Teja's Consejo was ineffective and polarised from 1995 to 2000 because the divisive Beatriz Silva was Junta secretary during that period (ibid.).

Another common complaint was that the Consejos had become 'municipalized'. The items discussed at meetings often addressed technical concerns about infrastructure, leaving little time to address the larger fiscal and political forces that constrain zones' agency. Without sufficient space to talk about larger issues, many councillors felt council meetings to be a waste of time (Canel, 2001).

Popular administrative participation continued to grow in spite of the 1993 reforms. The percentage of households that had used their local CCZ grew from 29 per cent in 1992 to 55 per cent in 1999 (Goldfrank, 2002). The aforementioned service gains can reasonably be attributed to the CCZs. Goldfrank (2002) writes that resident satisfaction with transportation, public lighting and street sweeping increased by at least 10 per cent in Zones 3, 7 and 9. In projects with larger capital costs, however, the city often proceeded without local input. El Cerro's Parque Tecnológica Industrial del Cerro was planned and built solely by the city and local entrepreneurs (Canel, 2010). An abandoned building that the La Teja Consejo wanted to turn into a community centre was leased by the city to a private company and converted into a banquet hall (ibid.).

Discussion and Conclusion

Three main conclusions emerge from the above. Firstly, decentralisation platforms are not always the product of expertise or ideology - they can simply be means to political ends. In the 1990 BPS bailouts and in the 1996 constitutional reform, the Blancos strengthened the power of the departmentos, where they were already dominant. Montevideo's radical FA leftists and Colorados sought to use decentralisation to mobilise the masses to serve party ends, a fact reflected by the Juntas Locales. Montevideo's decentralisation institutions became battlegrounds for party politics, alienating residents. The foregoing supports Stiefel and Wolfe's (1994) observation that participation can be the victim of both conservative forces ideologically opposed to decentralisation (the FA radicals) and of opposition parties intent on unseating the governing party (the Colorados).

From that follows the second conclusion: that interparty competition is a sufficient condition for institutional change in legislative democracies. At the national level, the legislature's lawmaking function allowed the Blancos and Colorados to build themselves up through strategic legislation. The Montevideo FA implemented devolutionary decentralisation in 1990 because it faced no significant pushback. By 1993, internal and opposition revolt had forced decentralisation to be recast in the mould of deconcentration. Consejos' lack of decision-making authority made political participation seem a useless exercise for citizens. Insofar as political disempowerment directly determines quality of life, the post-1993 decentralisation in Montevideo did not independently improve the lives of the dispossessed.

The power of political parties to change the institutions of decentralisation shows that deliberative participation is highly contingent on the constellation of political forces at a given moment. The national-level COSEDE gave its Blanco-dominated departments ('constituents') more influence over the institutionalisation of fiscal and political transfers to the departmental level, largely abandoned by the Colorados. The FA, the only 'game in town' in Montevideo in 1990, had free rein to devolve power to zones. By 1993, opposition forces once again had the political capital to dismantle the FA's policy. While COSEDE had won both material gains (annual fiscal transfers to departmentos) and political gains [departmentos negotiated as a bloc (Eaton, 2004)] for the actors that it represented, the decentralisation of post-1993 Montevideo performed well on the services front but failed to systematically involve the dispossessed in the political decisions affecting their lives. Decentralisation can improve the lives of the dispossessed when there is the political will to create institutions toward that end, but intense interparty competition makes the creation of such institutions a distant dream.

(1) This section up to this point draws heavily on Filgueira et al. (2002).

(2) This section up to this point draws heavily on Eaton (2004).

Works Cited

Araujo, M. C., Ferreira, F. H. G., Lanjouw, P. & Özler, B. (2008). Local inequality and project choice: theory and evidence from Ecuador. Journal of Public Economics 92, 1022-1046.

Canel, E. (2001). Municipal decentralization and participatory democracy: Building a new mode of urban politics in Montevideo City? European Review of Latin American and Caribbean Studies 71, 25-46.
- (2010). Barrio democracy in Latin America: Participatory decentralization and community activism in Montevideo. Pennsylvania State University Press. 260 p.

Dillinger, W. R. (1994). Decentralization and its implications for urban service delivery. The World Bank. 39 p.

Eaton, K. (2004). Risky business: Decentralization from above in Chile and Uruguay. Comparative Politics 37(1), 1-22.

Faust, J., von der Goltz, N. & Schloms, M. (2009). Promoting subsidiarity-oriented decentralization in fragmented polities - some lessons from Ecuador, in Altenburg, T. (Ed.), From project to policy reform: Experiences of German development cooperation. German Development Institute.

Finch, H. (1998). Towards the new economic model: Uruguay 1973-97. University of Liverpool Institute of Latin American Studies. Research Paper 22. Available at http://www.liv.ac.uk/media/livacuk/rilas/publications/paper22.pdf

Goldfrank, B. (2002). The fragile flower of local democracy: A case study of decentralization/participation in Montevideo. Politics & Society 30(1), 51-83.
- (2007). The politics of deepening local democracy: Decentralization, party institutionalization, and participation. Comparative Politics 39(2), 147-168.

Habibi, N., Huang, C., Miranda, D., Murillo, V., Ranis, G., Sarkar, M. & Stewart, F. (2003). Decentralization and human development in Argentina. Journal of Human Development 4(1), 73-101.

Herzer, H. & Pirez, P. (1991). Municipal government and popular participation in Latin America. Environment & Urbanization 3(1), 79-95.

Huntington, S. (1981). American politics: The promise of disharmony. Harvard University Press. 320 p.

Kim, S. H. (1992). The political process of decentralisation in Peru, 1985-1990. Public Administration and Development 12, 249-265.

Litvack, J., Junaid, A. & Bird, R. (1998). Rethinking decentralization in developing countries. The World Bank. 40 p.

McConnell, G. (1966). Private power and American democracy. Alfred A. Knopf. 397 p.

Oates, W. E. (1972). Fiscal federalism. Harcourt Brace Jovanovich. 256 p.

Rondinelli, D. (1980). Government decentralization in comparative perspective: theory and practice in developing countries. International Review of Administrative Sciences 47(2), 133-145.

Seabright, P. (1996). Accountability and decentralisation in government: An incomplete contracts model. European Economic Review 40(1), 61-89.

Stiefel, M. & Wolfe, M. (1994). A voice for the excluded: Popular participation in development: Utopia or necessity? Zed Books. 276 p.

van Ginneken, W. (2003). Extending social security: policies for developing countries. International Labour Review 142(3), 277-294.

Veltmeyer, H. (2007). On the move: The politics of social change in Latin America. Broadview Press. 209 p.

Winn, P. & Ferro-Clerico, L. (1997) Can a leftist government make a difference? The Frente Amplio administration of Montevideo, 1990-1994. In Chalmers, D. A., Vilas, C. M., Hite, K., Martin, S. B., Piester K. & Segarra, M. (Eds.), The new politics of inequality in Latin America. Oxford University Press. 644 p.