5. Decentralisation, Lobbying & Capture

On Accountability 

• The theme of the 2004 World Development Report is  summarised by its opening paragraph:
  
  

"Too often, services fail poor people – in access, in quantity, in quality. But the fact that there are strong examples where  services do work means governments and citizens can do  better. How? By putting poor people at the center of service  provision: by enabling them to monitor and discipline service  providers, by amplifying their voice in policymaking, and by  strengthening the incentives for providers to serve the poor."

Decision making not

Centralised bureaucracy (federal) engage in practices which do not keep the concerns of the people at the centre. 

More on Accountability 

Problems of accountability associated with traditional  modes of delivery involving centralised bureaucracies  include:

• cost padding, 

• service diversion,

• limited responsiveness to local needs,

• limited access, and

• high prices charged especially to the poor.

Similar to planned soviet economies (scissors crises) involving centralised bureaucracies where incentives are not aligned.

Remedies 

Many developing countries have thus begun to experiment  with initiatives to increase accountability of service providers by providing greater control rights to citizen groups, these include: 

• decentralisation of service delivery to local  governments,  

• community participation,  

• direct transfers to households, and  

• contracting out delivery to private providers and NGOs.

The programmes include a wide range of infrastructure services (water, sanitation, electricity,  telecommunications, roads) and social services (education, health, and welfare programs).

Countries where such trends (movements) have gathered momentum in span different continents:  

• Latin America (e.g., Bolivia, Brazil, Colombia, Costa Rica), 

• Africa (Ghana, Uganda, South Africa) and, 

• Asia (Bangladesh, Indonesia, India, Pakistan). 

On Decentralisation 

What is decentralisation?

• giving decision making power to local governments from central government 

The different aspects of decentralisation:

• • Political decentralisation:  local councils have full discretion to the use of collected taxes

  • Fiscal decentralisation: local council has right to collect taxes (adjusting tax rates) 

The associated costs and benefits:

Decentralisation as Fiscal Federalism 

Much of the fiscal federalism literature focuses on the economic efficiency of intergovernmental competition.

Based on the Tiebout (1956) circular model: different local governments (on differing nodes) offer different public tax-expenditure bundles,  and mobile individuals are supposed to allocate themselves according to their preferences.

• Therefore, people move to their preferred locations based on these policies. US federal tax system lightly based on this.

• Assumes people vote on their feet, but population are not that mobile when it comes to policy on tax. 

• Agrarian societies rellocate, but that is for seasonal changes resulting from joining their local economies to engage in trade, typically with their own produce.

The assumptions required for the Tiebout model are, however,  much too stringent, particularly for poor countries. 

Population mobility that enables governments in the Tiebout  framework to overcome the well-known problem of inducing  citizens to reveal their preferences for public goods largely fails  in most contexts. 

Decentralisation: Some Observations 

The trade-off between uniformity of service provision under centralisation with problems of uneconomic scale and cross regional externalities under decentralisation. 

The assumption of uniform provision under centralisation is neither empirically realistic, nor well founded theoretically. They assumed it is uniform where one size fits all which does not capture the true needs of the people. However, on a large scale when it revolves around the provision of public goods it may be more efficient due to scales of economies to prop up that one company/inter-governmental branch to manage and provide such goods.

Problems of externalities w.r.t. local taxation stressed by the  traditional literature are also not a practical concern. Why? 

Even if you give local councils control there could be spillover:

• Education: doesn't just benefit people in village/city, it may benefit the neighbouring villages/cities. Therefore, creating market ***(30 minutes)

Decentralisation in most developing countries generally involves delegation of service delivery systems to local  governments, without an accompanying devolution of  financing authority:

• Economies of scales typically arise from production, with distribution this is less of concern. If there is delivery being passed out to local communities ***

• Side-step issues which plague decentralisation by doing a partial-decentralisation (limited to not delivery, producing, etc), to exploit economies of scale, improve accountability and improve efficiencies at the local level.

Many of the programmes involve relatively few inter-jurisdictional spillovers, e.g., local water and sanitation projects. 

Economies of scale if significant tend to matter only with respect to production rather than distribution. 

So, production can continue to be centralised, and local governments may procure the service and decide how to allocate it within their respective communities. Therefore, decentralisation does not involve any of the welfare losses stressed by the traditional literature.

The major concerns frequently expressed with  decentralisation are that local democracy may not function appropriately, thus limiting accountability of local government officials or community leaders.

With limited political contest-ability of local elections, leaders may be susceptible to capture by special interest groups, slacken effort to improve public services, or be incompetent (lack of training), without facing any risk of losing their positions. 

In that case accountability, efficiency and equity in service delivery may worsen under decentralisation. 

Policies must be serving the broader people

Capture and Governance: Bardhan and Mookherjee (2000) 

The literature on political economy has several analyses of capture of the democratic process by special-interest groups. 

Special interest groups formed by elites 

This is a study on the relative capture at central and local levels of government. 

• One view is that the lower the level of government, the greater is the extent of capture by vested interest (“local elites”).  Then all the gains can be turned on its head.

• The other view stresses the advantage of decentralising delivery mechanisms to local governments with access to superior local information. This provides for the people.

A Model 

There are n districts each with an identical number of voters. 

The voters are divided into three classes: poor (p), middle income (m) and rich (r).  

Districts differ in demographic composition across the three  income classes. 

• A fraction 𝛼c  of voters in class c are informed (politically aware) and vote for different parties partly on the basis of the levels of welfare they expect to achieve under their respective policies. 

• The rest (1-𝛼c ) are not informed.

The Voters 

Political awareness is closely related to socioeconomic position and education level, so 𝛼r > 𝛼m > 𝛼p

An increase in the fraction of the population that is poor will accordingly imply a lower fraction of informed voters in the population as a whole. 

An informed voter votes on the basis of expected utility from either party’s policy + ideology. 

An uninformed voter is swayed by (the balance) campaign spending +  ideology. 

These are probabilistic voting models. Both behavioural, being informed but not necessarily possessing all of the information available in the market. An uninformed voter behaves rationally alongside the informed voter. 

Political Parties and Lobbying 

In the Downsian tradition, parties announce policies prior to the election and are assumed to commit to these once elected. This commitment is rationalised by repeated games, as next elections come and they did not fulfil promise then they would not be elected again. This may result in all parties deciding to not adhere to their promises, but then new parties would come (entry into a market, SPNE equilibrium)

There is a single organised lobby, composed only of the rich who are able to (can afford to) donate.

A given fraction of the rich contribute financially to the lobby, while the remaining members of this class free-ride. The rich fund the politicians to win the swing voters, but of course would like something in return, wanting to know what the policies would be, and in favour of them.

The lobby contributes to the campaign finances of the two parties, conditional on their policy platforms. 

The Parties’ Objectives… 

Given these contribution strategies, each party selects a policy to maximise its probability of winning the election. This is where the Subgame Perfect Nash Equilibrium is introduced in the model.

Each party will try to maximise a weighted sum of the average:

• utility of the informed voters and,

• the campaign spending (for  winning the uninformed voters’ support). 

In the case where all voters are informed, the weight on campaign spending is zero. The only thing the campaign spending is doing is reaching to the un-informed and becomes less worthwhile. Diversion from ideal, efficient policy stems from the uninformed.

Policy biases arise from the existence of uninformed voters.

The Factors Determining the Extent of Capture 

• Interest-group cohesiveness: represented by the fraction of  the class of rich citizens who contribute to their lobby. There will be free-riding even amongst the rich. If cohesiveness is not very high then the lobby may well collapse. The average value of 𝛼 is high in a constituency, it is less prone to capture.

• Average level of political awareness: which is increasing in the  fraction of uninformed voters in the district. 

• Disparity in awareness levels across classes: represented by  the fractions of voters αc, in different classes who are informed. If the average value of 𝛼 is high in a constituency, it is less prone to capture.

The last two factors explain why capture increases with illiteracy, poverty, and inequality. 

The Situation at the National Level 

First Order Importance: substantially large, most important factors in a decision, system, etc.

Second Order Importance: less critically important

Suppose that the policy space is the same as at the local level.  

Assume that owing to the lack of suitable information regarding differences across districts, national governments are constrained to select the same policy across all districts. This results in uniformity. 

With decentralisation, this constraint no longer operates, allowing greater "flexibility" with respect to local conditions. 

The objective of party k will be to maximise Vk  which aggregates vk, its objective in local elections across the districts. In every constituency is looks at the vote share, then they add it all up to determine to what extent you are politically relevant.**** 70 minutes

Basic Results 

Suppose all districts are identical (not the case in reality):

• Voters are better informed (information is greater) at the national level owing to greater media attention (everyone is aware of the politicians), 

• Or the rich are less well-organised at the national level (owing to greater size and heterogeneity of the group, or larger communication and coordination costs). 

• If uncertainty is greater at local level, good news for poor, less capture possible. Local elite face a gamble as they have to equally pay between the two candidates in order to guarantee success of their own interests. Makes a case for decentralisation. Amount of electoral uncertainty is greater creates a bigger case for decentralisation

Each locality is different, this encourages us to decentralise.

Suppose that electoral uncertainty is greater at the local level. This will imply less capture available at the local level. Dominant party is less likely to win at the local level, thus reducing the incentive of the lobby to contribute to its campaign  funds.  

Extensions 

Consider a two-party system at both levels, but where districts vary with respect to inequality (inequality distribution is widened)

This tends to increase capture in high-inequality districts, owing to the higher fraction of uninformed voters in such districts, creating  disparities in the effectiveness of campaign funds across districts.  

These disparities imply that in a national election parties will bias the allocation of campaign spending in favour of high-inequality districts. Local elections in low income*** 80 minutes

In this case capture at the national level equals the highest level of capture across all local governments.  

Inequality is a huge driver as to whether you decentralise or not.

Conclusion 

The relative proneness to capture of local governments depends on a  multitude of diverse factors.  

Some of these provide support to the presumption in favour of  greater capture at the local level, such as greater cohesiveness of  interest groups and higher levels of voter ignorance at the local level. Extent of electoral competition

But we also identify a number of other determinants of capture  which pull in different directions. These include the relative extent of  electoral competition and intra-district inequality. 

This creates the need for empirical research to identify the nature of  relative capture in any given setting, in order to appraise the  potential pitfalls of decentralisation.