Home | News | Discussion | FAQ | Federal Model | Papers | Seminars | What's New

Paper on Distribution of Public Revenue at Central/Federal and Local Levels of Administration

by Pancras Kaboha, Senior Lecturer, Department of Philosophy, Makerere University

The Distribution of Public Revenue at National and Local Levels in a federal or decentralized State:

1. Factors Determining Distribution of Revenue

In being able to arrive at a workable formula for the distribution of national revenue in a federated state, it is necessary to look at two things:

a) The level of responsibility given to the Central Government as compared to the local governments; and
b) The capacity of local governments to raise revenue in their areas of location.

A: Level of Responsibility

As regards the division of responsibility between the local and central governments, various formulae have been used including the model practiced by Uganda following Schedule 9 of the 1962 Constitution. Normally the central government retains control of import and excise revenue as well as the higher levels of the administration of justice, defence, security, and foreign affairs. There is a sharing of responsibility in all the other areas of government work and it is the extent of this responsibility that determines or should determine the level of revenue that should be retained at the different levels of government. To cite a few examples in the clarification of this we have cases of internal security where in Uganda and Kenya we have had central police, and local administration police which complement each other and get their pay and facilities from central and local administrations respectively. In the field of education, central government and parents have tended to take all the responsibility but local governments can participate in contributing to primary schools or even for the upkeep of students from their areas who happen to be in higher institutions of learning which take students from all areas of the country and can only be supported centrally. In health, local governments can (as used to be the case in Uganda), run dispensaries which are normally located at sub-counties while district hospitals can remain the concern of the central government, but local district administrations can contribute to these depending on the amount of revenue available to them and take part in their administration. In agriculture, local administrations can stop at sub-county, county or district level in contribution according to what has been the administration formula used in the decentralization of power. Agricultural assistants, or Veterinary assistants and all officers of this grade were normally the people who fell into the grade of responsibility for the districts, but with the recent decentralization statute (Decentralization Statute 1994), it appears that nearly all district officers in Uganda will be answerable to the local administration and the local councils, and therefore a more thorough overhaul in the division of responsibility and finance is going to be necessary.

When federal units are considered in terms of decentralization, it would appear that in most of Uganda, these units would be districts or a combination of two districts; and the fairly large federal area would be Buganda. For the purpose of this paper, therefore, Buganda will be considered as a super "district" the type the delegate for Busiro had referred to in the Constituent Assembly (CA Hansard), and when I say "district", I will also mean the super district of Buganda. Because of the current focus on the district in Uganda, we shall consider it as our focus of federalism or decentralization for the sake of this paper and will thus discuss the devolution of power and the shared financial and administrative responsibility at that level, regardless of the size of the district.

On deciding the levels of responsibility, we can say that if the districts are to be truly federal units, then they should have the responsibility to appoint, pay and discipline all district government officers serving in the area and the Public Service Commission should enable them to do this. However, this is not as straightforward as it sounds because security organs like army barracks, intelligence units, police, immigration and customs units (at border points and airports) have to remain under central government control.

B: Capacity of Local Governments

After formulae and percentages of contributions have been worked out, there will be districts, because of the limitations of manpower (e.g. Kotido, Kalangala and Bundibugyo) or revenue generating resources, which will lack sufficient revenue to pay for their maintenance as a part from their development. Such a situation will require grants from the central government to even up the shortfall and even to mount development projects which would enable the district to pay itself in future. Even if formulae of revenue sharing are worked out nationally, such districts would have their contributions either suspended or more than compensated for by central governments "grants in aid" as the colonial terminology used to refer to them. These grants have unfortunately been used by some of the campaigners for federal status in the belief that federal arrangements would put a stop to them. However, federal governments all over the world have a money sharing formula which takes into account such grants for areas of the country which are less developed than others. This is a bone of contention which federalism does not solve, but what it does solve is to leave the decision of how to spend the remaining money (after the central government cut has gone off) to the local authority which sets its own priorities. Apart from the financial capacity of a district to look after itself there is also the planning capacity of the district whose plans must also be synchronized with those of the central government. There are centrally planned and distributed inputs like roads and power supply and the locally planned projects have to take into account the infrastructure that these services provide as well as the training centres that provide the necessary manpower. No district can afford to plan as if it is an island in the nation and though its federal status can allow it independent planning, its plans must be synchronized with the plans of the other districts and the central government so that the sharing of services can be rationalized, and duplication can be minimized. The capacity of the members of a district or federal unit to think nationally and plan along with the rest of the nation is important especially in an underdeveloped country where so many resources are centralized and have to be shared.

2. Possible formulae for sharing of revenue in a decentralized or federated State:

These fall into the following categories:

a) Maximum devolution of power and local centered government planning and financial control;
b) Moderate devolution of power and shared government planning and financial control; and
c) Minimum devolution with centralized planning and limited control of local finance comparable to a unitary state.

A: Maximum Devolution

Under this model, the local federal units are practically independent of the centre and yield only defence, foreign affairs and such other functions as cannot be rationally decentralized. The nearest approximation to this are the Swiss Cantons (in Switzerland), whose independence is proverbial. The twenty two cantons which form the confederation of Switzerland have extensive local powers covering most of the essential services and thus keep a substantial percentage of the taxes they collect for local government. In spite of having a lower house (a national council of state) in which houses all the cantons are represented, no matter of real national importance can be legislated without consulting the full cantons, and serious changes in national policy require a referendum.

The United States do not fully fall in this category but certain aspects of their federalism are extreme partly due to the sheer size of the nation and the independence of the people. These aspects include certain laws which may change from state to state, and their penal code which does not always carry the same kind of punishment for similar crimes. The most extreme examples of this is the death penalty which obtains in certain states (e.g. California) but not in others. State governors and their assemblies maintain a considerable degree of independence from the centre, and in matters of finance plan and spend their funds freely after their contribution to the federal government. State and city police is independent but acts in collaboration with the Federal Bureau of Investigation which is a national or federal agency. Federal funds do come in to help in areas of health, education and welfare as well as in communication and other services, but the bulk of the expenditure is faced by the state which retains the bulk of the revenue. Maximum devolution though possible in a small state, is more practicable and ideal in a large nation where the proliferation of authorities and powers would not stifle work or progress. Nations of the size of the United States or the former Soviet Union are the size of nation that could easily tolerate maximum devolution and decentralization so long as there are checks and balances from the central authority. Federated bodies or nations always live with the threat of units breaking away, not only did the United States fight a civil war to keep the union together, but once the Lukiiko of Buganda did "order" the central government of Uganda off Buganda's soil (Akena-Adoko: 1966 Crisis). Although this latter was a mere act of bravado which Buganda did not have the means of enforcing, it provides a background why people feel insecure about a very "loose" federation where too many powers are decentralized. We also have examples of what happened at the failure of communism in the former Soviet Union. All the former federated republics broke away and the very loose commonwealth of independent states never really took off. Very loose associations which are created for mainly economic reasons have been failures especially in Africa. Europe has had many struggles with the European community which is eventually aimed at a political union. The Organization of African Unity which has the same aim is also starting with economic associations which on the whole have failed to take off. After spending time and money negotiating and establishing the PTA COMESA and other bodies, now Uganda, Kenya and Tanzania are trying to awaken the dead community. It would appear therefore that the successful union and maintenance of states is based more on political than economic organs and devolution of political power must not reach a level which destabilizes the state. Extreme devolution as a model carries this risk and is therefore, desirable in very large states, which have very highly developed central institutions and sense of nationhood. It could prove disastrous in a nation where people identify themselves more with their tribes than with the nation state.

The sharing of revenue under maximum devolution is often commensurate with the level of responsibility. Since so much has been devolved, all education all health (except referral and specialist services), all internal roads, all federal unit (district) security, courts, prisons and other services are met by the federal unit or district and not the central government. In a case like this, up to eighty percent (80%) of the district revenue can remain to carry out these services and federal grants can still come from the centre to top up services that the district must provide and the few central government agents who must be provided for. Such local administrations are almost as state within a state and recruit, pay and retire their own civil servants. They have their own local bodies for most government work which are not answerable to the central government.

B: Moderate devolution of power

Under moderate devolution of power, there is a sense of shared responsibility for all national government work except for areas like defence and foreign affairs which is central government work, and for government work at the lower units like feeder roads between parishes and other work at the small sub-units of districts which is local government work. There is clear delineation and division of areas of work and responsibilities falling in both categories of central and local government, but there is also a lot of co-operation in this category as there is a lot of meeting each other half way in areas of responsibility and helping each other achieve common goals.

The level of distribution of responsibility varies slightly from country to country as the needs of each country differ but we can develop a Uganda model based on our peculiar circumstances and the history of sharing we have experienced. A moderate level of responsibility sharing requires a division of revenue that does not divert far from the mode 50% - 50%. Depending on the circumstances within the country, this can change to 60% - 40% or 45% - 55% with the higher percentage pending to remain in the local area as it may not have many sources of alternative revenue.

In the case of Uganda, where we have been operating under minimum devolution of power (to the detriment of the countryside), we have to give more funds to the local authority to facilitate the development of the rural sector which has been greatly neglected since government has been over-centralized since 1967. The neglect and underdevelopment of the rural sector has also resulted in the over-centralization of industry and development in the chief urban areas. As a result, nearly 70% per cent of the national revenue is collected from the excise and import duty as well as sales tax (Presidential Address on Budget Day (June 1994)). The urban industrialists and importers are responsible for the greatest part of the national revenue and this makes it necessary for at least half of this urban revenue to be ploughed back to the rural areas, where over 80% per cent of the Ugandan population lives, according to need. If it is not ploughed back services will not be provided to the rural sector, there will be no development or infrastructure to attract industry, and there will be mass migration from the rural to the urban areas. So when considering our sharing of revenue under our moderate level of sharing, we must be aware that practically, all local governments not located in the big industrial areas like Kampala and Jinja will need grants to top up their revenue requirements.

So what would be the responsibility formula suitable for Uganda under moderate devolution model? I would suggest that we actually look at all government work practically ministry by ministry and see at what level ideally local government can come in. If we look at education, I would suggest that local governments take up all primary and nursery schools which are not privately funded (and founded) and budget for their maintenance, construction of new ones and remuneration for the teachers. This will build up concern for the lower forms of education in the rural areas and uplift their standards to match those of the bigger urban areas. The local government should also be responsible for the appointment, discipline and remuneration of the education officer or officers in charge of primary or basic education. Secondary education would not be ideal for the local authority as many of the students come from other localities and authorities and local interest is not necessarily paramount. There is no reason or cause to stop the local authority from providing bursaries to its own native sons and daughters even to tertiary institutions which are normally national. The local authority may, if it wants, build secondary or tertiary institutions which it can control directly.

When we consider the Ministry of Health, we can say that the local authority may stop at the local dispensary level and concentrate on spreading these to every parish rather than having them at the sub-county level as has been the case in most districts, or the local authority may concern itself with the district hospital as well which has been practically just the one hospital per district. As far as I am concerned, health services have been very poor in Uganda and the central government, has dismally failed to maintain the district hospitals. Given this circumstance, central government should provide grants to local authorities to maintain these district hospitals and provide personnel to man them who can be paid locally. (Junior staff can be locally recruited while the Public Service Commission can second doctors to the Local Service Commissions to work in district hospitals). However, national health policy should be maintained through co-operation of officers and the local councils, plus constant communication from the Ministry headquarters. Central medical stores can have local district equivalents where drugs go and can be subjected to local control by the community affected by shortages. Training centres should remain with central government control. Health assistants, visitors, nurses and employees of this middle cadre should be directly appointed by or seconded to (by the Public Service Commission) the Local District Service Commission.

The Ministries of Agriculture, Veterinary Services (animal husbandry), Forestry, Culture and community development, who have middle cadre employees who are below the level of district officers (e.g. agriculture and veterinary assistants) can divest themselves of these officers to the local authorities. District Officers can work as co-ordinator between the central government and the local governments. Their payment (District Officers) can be dual in that they can receive fixed pay from the centre and allowances from the district for their coordination work. The Ministry of Public Service can divest itself of a lot of its centralized work to district Service Committee who can take care of all the workers below the District Officer representing a specific Ministry. In those Ministries where we still have provincial or regional officers, these can represent the central government and be under the Public Service Commission and the District Officers can be under the District Service Commission.

These Options can remain open but a certain amount of control by the District Service Commissions is needed over all district officers to have harmonious working conditions for all, but also a certain answerability to the central government must remain in place at the level of the district head.

Ministries like Information, Labour, Sports, which have only district officers should be appointed and paid by the centre, however, they should sit with the local district council and be answerable for their actions to it.

The Ministry of Justice can be directly responsible for the Chief and District Magistrates but the lower Magistrates can be appointed and paid by the District Service Commissions but their supervision like that of secondary school teachers should be left to the Judicial Service Commission and the Teachers Service Commission, unless these are decentralized as well. As a matter of fact, primary school and nursery school teachers need a district Teachers Service Commission.

Police and prisons can either be left at the centre but I would suggest decentralization below the level of district headquarters, so that policemen serving at the Saza level and local prison Commanders below the District Commander himself should be answerable to the District Service Commissions. To achieve this effectively, District Service Commissions should include retired officers from these specialized services to advise them as to their mode of operation.

The model suggested here is intended to go along with the moderate decentralization or devolution model which does not over-centralize or over-decentralize power and responsibility and therefore recommend a division of revenue not far from parity. I shall deal separately with the role of the ministry of Local Administration and the President's office in the conclusion to this paper because these bring about the points of most confusing and mixing up of roles in our current decentralization exercise (Decentralization Statute 1994).

C: Minimum Devolution Model

Since the abolition of district councils in Uganda and the centralization of control of all the work of the Ministries in 1971 by the Idi Amin administration, we can say Uganda provides a good level of the minimum devolution model. The chief characteristics of this model is centralized planning, maximum control of finance and administration from the centre, and the total dependence of the local administration on the centre. This is the worst aspect of unitary states and is recommended for very small countries with a homogenous population and a very good system of communication and a well developed service infrastructure which negates the need for several local authorities. However, Uganda had not reached this state and the overall result was an inefficiency which resulted in the deterioration of the existing infrastructure.

The coming of the Resistance Councils (RCs) did improve the aspect of local representation but did not deal with the financing aspect of local administrations. As a matter of fact, corruption spread because of remote control from the centre. When mentioning Ministry of Finance representatives in districts, the Auditor General's department needs a strong mention, but the districts themselves might need an independent audit team given Uganda's peculiar financial control situation. The financing and altered administration of the district is what necessitated the current decentralization exercise with a number of pilot districts being experimented on (Decentralization Statute 1994). Most Ugandans have agreed that decentralization is overdue as a solution to some of our ills but the model that the government has started with has left many dissatisfied; and that is what I shall devote my conclusion on today. Some people have argued that maybe the model was introduced to pre-empt the federal demands of Buganda. Whatever the case, one got the impression that quickly as the government introduced the bill, its devolution of real powers did not take place, the government was going very slowly about devolution of power. A good example of how little of real power local authorities have I shall talk about in my conclusion.

Conclusion

Recently, the Kampala City Council, one [of] the mightiest of our urban and local authorities, awarded the administration of a Kampala taxi park to UTODA a taxi company. The Minister of Local Government, using the powers granted to him by the statute, revoked this decision. Any decision taken by an urban or rural authority stands in this jeopardy of approval or disapproval by the Minister. The constitutionalization of federal institutions to which real powers have been devolved does not envisage this interference by the Minister or else the effective ruler of all the local governments shall be the Minister of Local Administration.

In a properly federated or decentralized country, there should be no Minister of Local Administration. This concept came from the British concept of the Minister for Colonial Affairs. When any colonial body wanted to have anything to say to the British government, it had to go through this person who was the "boss" of the colonial governors and whose office distanced the colonies from the British public and government. The Minister was a dictator.

Colonial bodies were not of the status to be heard in the foreign office (though Buganda tried (Desecration of My Kingdom, Mutesa. The Mind of Buganda, David Low)), so their affairs remained in the colonial office. Here again affairs coming from any area of Uganda will first be handled by the Local Administration Ministry. The same affairs can go direct to parliament or even to cabinet through other Ministries, the central government representative reports directly to the President, so what is the purpose of the local government Minister except as a "stopper" employed by the cabinet team? Once federated or decentralized areas are given their own administrations, they can deal directly with any Ministry and government body and do not require the bureaucratic delay instituted in the body of the Ministry of Local Government.

What I see as necessary is for the Constituent Assembly to discuss this matter afresh and not basing themselves on the half-measure experimental decentralization statute now in place. Since there is no disagreement over the need for decentralization what is needed is now the final model which can be constitutionalized for posterity. This matter needs to be handled seriously and expeditiously in the plenary session. A further committee state may be necessary to finalize it.

What we have now is still a minimum devolution model in many aspects including that of the devolution of power and financial administration. The role of the Central Government Representative (C.G.R.) is still ambiguous in this context and the necessity of his existence is not clear.

If district administrative officers and especially the executive officer and especially the executive secretaries exist, they can take over the role of the Central Government Representative and can be employees of the District who co-ordinate with the President's office.

The presence of the Central Government Representative hovering above whatever the district councils decide as well as the Minister of Local Government make nonsense of serious devolution of power.