Previous Page Table of Contents Next Page

Chapter 5: The reform process: subsequent stages[50]


In the last chapter we started to examine the reform process by considering what could be done to improve the efficiency of agricultural service delivery without actually decentralizing the services. We argued that in certain circumstances quite a lot could be done through improving their internal efficiency and improving the competitive environment in which they operate. In this chapter we assume that the decision to decentralize has been made. The questions that this raises are: What institutional or organizational changes are required? How should the reforms be phased? Is the reform process working effectively? Each of these questions in turn raises a host of issues and we provide guidelines to assist this analytical process.


5.1 What institutional or organizational changes are required?

5.1.1 Introduction

In Chapter 4 we argued that it is appropriate to consider the possibilities of internal reform of the organization to improve its efficiency and the strengthening of the competitive framework within which it operates before actually moving to decentralization. At this stage the actual form of decentralization needs to be decided. In Chapters 2 and 3 we considered some of the theoretical and practical issues relating to this choice and in Section 4.2.2 we discussed some of the political implications. We now recap some of the economic and administrative factors that can influence this decision.

Criteria for framing questions

Decisions about decentralization of agricultural service functions need to take into account major policy objectives such as efficiency, accountability, transparency and equity. Four major general criteria are relevant in framing the questions that need to be answered in making these decisions:

5.1.2 Would deconcentration be appropriate?

Conditions for deconcentration

There are several factors that might favour deconcentration of agricultural services. It might be appropriate where most or all of the following conditions apply:

The questions that need to be addressed when contemplating deconcentration mainly relate to the competence of the existing field units. For instance:

Some personnel issues also need to be considered, such as:

Furthermore, deconcentration may also influence the quality or the quantity of the service being offered, particularly if cost recovery is being introduced.

Obviously, other questions will be raised depending on local circumstances and this checklist will need to be expanded or modified to take these into account.

5.1.3 Would delegation be appropriate?

Conditions for delegation

The situations that might make delegation the appropriate choice on economic and administrative grounds would be those where most or all of the following conditions apply:

Questions that would need to be raised when considering the delegation of responsibilities for service delivery include:

5.1.4 Would devolution be appropriate?

Conditions for devolution

Devolution of service provision to local government administration would be appropriate when most or all of the following conditions apply:

Many of the questions that need to be addressed when considering devolution are similar to those already considered regarding deconcentration. These include:

As with other forms of decentralization, devolution is likely to raise questions about the future employment of existing personnel that will need to be addressed.

5.1.5 Would partnerships be appropriate?

Participation with CSOs

Decentralization of service provision to CSOs serving communities or groups below the local government level, perhaps through partnership programmes, would be appropriate when most or all of the following conditions apply:

The role of government in partnerships with CSOs

Although it is preferable for government to work with existing organizations, there are several measures that government in partnerships with CSOs can take to strengthen their performance. Broadly speaking the function of government may include the following:

Delegating responsibilities to CSOs

When contemplating delegating responsibilities for service delivery to CSOs, several questions need to be addressed regarding CSOs and the government’s intended relationship with them. For example:

5.1.6 Would privatization be appropriate?

Private sector alternatives

Where the public sector has been producing ‘private’ goods it may be relatively easy to decide on the most efficient post-reform organizational form providing there are economically and financially viable methods of production available. For example, where there are no economies of scale and risk levels are low, there is unlikely to be any justification on cost-effectiveness grounds for continued central government production. In these situations private individuals or organizations such as small private firms or cooperative organizations can instead undertake production. This also has the potential benefit of spreading ownership and possibly creating a competitive environment.

Economies of scale

However, as economies of scale become important and/or the degree of risk increases, the minimum size of organization may need to increase substantially. If private sector alternatives to parastatal production are being considered, questions of obtaining adequate financing then arise. In some circumstances this may limit the range of owners to larger companies using equity finance from domestic and/or foreign investors. In some circumstances there may be political resistance to overseas involvement of what are seen as ‘strategic’ sectors of the economy. There are, however, more innovative forms of financing which allow for a range of ownership forms incorporating a wider spread of stakeholders. These are discussed in more detail later.


Potential problems of collusion between a limited number of competitors also have to be considered in these situations. For traded goods these can often be overcome by liberalizing trade, but for non-traded goods and services the government may need to consider strengthening its competition laws and/or ensuring that a variety of types of organizations are encouraged to provide the service. There are circumstances where central government may be justified in retaining control and/or ownership of part of the parastatal’s assets as a means of generating competition and providing the government with information for regulatory purposes.

Natural monopoly

In some circumstances the reason for public provision was the existence of a natural monopoly arising from substantial economies of scale relative to the size of the market. In these situations the two main alternatives to continued parastatal provision are through the use of management contracts to control the assets (see later) or through sale of the assets to the private sector. Both routes harbour risks relating to the government’s inability to effectively monitor and regulate the situation and avoid its own exploitation and that of consumers and these problems need to be confronted.


The reform of public sector organizations producing services with ‘collective’ properties can prove more difficult if the ‘free rider’ problem prevents a privatized organization making a reasonable rate of return on its capital. Agricultural research organizations are a good example of this situation. There are, however, potential solutions. One is to ‘internalize’ the externality, for example by devising organizational structures so that all potential free riders take a share in the ownership of the organization and hence in its success or failure. Another is for the government to accept that the externalities provide a positive benefit to certain groups in society and to financially compensate the organization for providing this public service.

Yet again, appreciating the ability to disaggregate ‘provision’ into its component parts markedly widens the potential organizational forms available when restructuring occurs, especially when one is dealing with ‘collective’ and/or ‘merit’ goods. This approach allows the comparative advantages of parastatal and non-parastatal organizations in providing different functions or various combinations of functions to be analyzed. On cost-effective grounds one might expect the role of the parastatal to change depending on:

Paradoxically however, even if the private sector is strong it requires effective government machinery for financing and regulating the provision of public and merit goods before the role of agricultural parastatals can be fully replaced.

Divesting the ownership and/or control of parastatal assets

If the examination of the relative roles of parastatals and the private sector leads to a decision that the role and/or ownership of the parastatal should change then the choices are to reduce:

Each of these strategies could be used in isolation or together. For example, it may be decided to recapitalize a parastatal and improve its performance as a prelude to introducing competition or to selling the assets. Again, the relative merits of these different approaches are likely to vary according to different circumstances and many of the factors influencing this choice are discussed below.

The prelude to divestiture

From a legal and administrative viewpoint, divestiture of assets can be an extremely complex procedure and a failure to appreciate this can lead to frustrations with the pace of divestiture and/or ill-advised privatizations being undertaken in haste. As Adam et al (1992) comment:

“it is one of the ironies that the management of privatization can in fact involve an extensive call on the time and resources of the public sector. This is manifested in a variety of ways, and governments have generally underestimated the extent to which the details of individual privatization transactions have to be dealt with before asset sales can go ahead.” (p.51).

Guislain (1997) provides a detailed guide to the many legal, institutional and management factors that need to be considered before a successful privatization can be implemented. Only a brief indication of these questions can be given here. In some countries the constitution of the country may need to be modified to allow privatization in general, or of certain types of activities. Similarly the current constitution may prohibit foreign investment in certain sectors such as land ownership. Again, in some countries prior approval of parliament may be required before the privatization process can proceed. International laws and treaties may also affect some aspects of intended privatizations.

Who owns the assets?

As privatization entails the transfer of ownership of certain assets from a public entity to a private one, clearly defined private property rights are of vital importance (Guislain, 1997). It is equally important to establish clearly who owns assets that are to be privatized and who has the legal right to agree to the divestiture of assets. This particularly applies to assets that have been confiscated, expropriated or nationalized in the past. In some instances there may be more than one group of stakeholders with some form of claim on assets. These claims have to be investigated and settled before privatization can occur. Again, this may require legislation to solve seemingly intractable problems. Legislation to ensure a level playing field and to avoid anti-competitive practices may also need to be enacted.

Do laws need changing?

Guislain (1997) also demonstrates that contract law, company law and shareholders’ rights, accounting law and laws relating to liquidation and bankruptcy may all need to be reviewed and strengthened to deal with successful privatization. Securities legislation may also be required to encourage capital market development and share dealing. Similarly, financial legislation dealing with banking laws and regulation, collateral and sureties, credit, leasing, insurance, etc. may all need to be revised to allow successful divestiture and private sector development. Taxation law, particularly as it relates to parastatals, may need to be clarified.

A whole range of social legislation may also need to be addressed. This may include labour legislation, pension provisions and social safety nets. Parastatal employees may have extremely favourable employment terms such as enhanced redundancy and pension rights that private sector employers would be unable to afford. Legislation may be required to change these terms before the parastatal is divested. There may also need to be pension fund reforms for workers who are to be transferred from the public to the private sector.

There are also institutional and management issues that will need to be resolved. In many instances parastatals may need to be financially restructured before divestment of assets occurs. This may include cleaning up the balance sheet, writing off or taking over part of the debt, renegotiating credit and loan arrangements with commercial banks and donors, setting up suitable financial accounting systems, etc. If the existing management are opposed to privatization, or are unable to adapt to a commercial environment, it may be necessary to replace them as a prelude to privatization.

Is the current legal status OK?

The legal status and organizational structure of the public sector organization may also need to be changed. For example, divestiture through a sale of shares will entail a change in status to that of a company organized under private law (Guislain, 1997). It may also be decided to break up the assets to encourage competition or to make the parastatal easier to sell. In some cases liquidating the parastatal and selling the assets may prove to be the easiest means of divestiture.

Even when careful consideration has been given to these various legal, institutional and administrative aspects, experience of actual divestitures, and learning by doing, may call for amendments to the procedures. It is thus very important to build sufficient flexibility into the original proposals that this can be done with the minimum of inconvenience and delay in the procedures.

The sale of assets by the state

The divestiture of assets by the state is increasingly viewed as the most desirable method of parastatal reform by many donor agencies. Indeed, they would view any attempt to improve the internal efficiency of public sector organizations as a waste of time. This reflects a growing view that the governance problems associated with parastatals are such that good performance is impossible, and also that the state should withdraw from as many activities as possible. It also has the advantage that it may yield a one-off fiscal gain from the sales proceeds, although Guislain (1997) cautions that the net proceeds of divestiture have been disappointingly low in many situations. The actual sale of assets is a political as well as an economic process and its successful implementation involves many facets and detailed procedures that need to be fully considered before privatization occurs. Some of the issues that need to be addressed are examined below as a series of questions.

How should the assets be valued?

One of the many difficult issues is to establish the value of enterprises or assets that are to be divested. If the divestiture occurs through a transparent, competitive process with well informed buyers then a ‘true’ market value would be established through the bidding process. However, even with well-informed buyers there can be large uncertainties surrounding the future prospects of the enterprise so that the valuation made at the time of sale may appear to be wildly inaccurate after only a short period of time. Often the sale is made at a pre-determined price and it is in these circumstances that accusations are frequently made that assets have been disposed of too cheaply to favoured purchasers. For this reason it is desirable that independent and qualified experts using a range of generally accepted valuation principles carry out valuations, rather than the existing managers or others expected to benefit directly from the sale.

Who will be allowed to purchase the assets?

As with earlier strategies, another political consideration is the range of actors that are permitted to engage in the privatization process. On efficiency grounds it might be argued that the range should be as large as possible, but in reality some bidders may have better access to financial resources than others. Political and social factors also have to be considered. For example, potential purchasers may be restricted to the existing management and/or workforce and/or clients. This is most likely when these groups represent a strong constituency and the government wishes to protect their interests. Alternatively, the government may be concerned to establish a local capitalist or rentier group and divestiture may be limited to national organizations only, or to alliances of one or more of the stakeholder groups. However, there are several instances where the inability of local individuals or organizations to raise finance to purchase privatized assets has seriously retarded a privatization process. On the other hand, ownership or control may be extended more widely to embrace foreign organizations and multinationals either in their own right or in alliances with national firms and/or stakeholder groups. The advantages of this wider constituency are that it widens the range of potential bidders and includes those who are likely to possess the most resources.

What is the most appropriate method of sale?

Problems can also arise in deciding the appropriate public or private method for the sale of assets, although the type of sale and the range of permitted purchasers tends to be linked. As indicated above, lack of local finance can be a serious impediment to the sale of agricultural parastatals to nationals of a country. Guislain (1997) indicates that two general methods of overcoming this are through deferred payment schemes (seller financing) and credit provision (bank financing). Public flotations of shares in parastatals require a proper functioning stock exchange (Bennell, 1997). In countries without stock exchanges, privatization trust funds (PTFs) can be established that allow governments to keep parastatal shares in trust until such time that they can be sold to the public. In their survey of seven developing countries, Adam et al. (1992) stated that most countries used public share issues through a formal stock exchange for major divestitures. However, the single most common method involved the direct sale of an enterprise to a domestic or foreign purchaser, or to a joint-venture consortium of both. Bennell (1997) surveying divestiture in sub-Saharan Africa states that there are three basic types of private sales:

Leasing arrangements

An alternative to outright sale is to lease the assets to a private organization for a stated period, normally for a fixed rent. This may be an appropriate strategy where the government is uncertain whether it wishes to lose complete control of the assets or their output. For example, the conditions of the lease may specify that certain services must continue to be provided. Another advantage of leasing over outright sale is that this allows the lessee to finance only working capital requirements rather than having to find finance to purchase fixed assets. This can increase the number of competitors willing to bid to use the assets by lowering barriers to entry substantially. It also allows the government to find new lessees if the terms of the lease are not met. The most effective method for initiating the lease is probably by competitive tender. Long term leases, and an ability to bid for subsequent leases, are likely to encourage more investment by the lessee and act as a disincentive to mining the assets, particularly towards the end of the contract.

Management contracts

Contracting out parastatal management to the private sector (Box 5.1) can be used as a way to improve parastatal performance or to prepare it for full privatization. In the agricultural sector, the World Bank survey (1995) found that management contracts are sometimes used for running plantations and agro-processing facilities for products such as tea, rubber, sugar, etc. There tends to be a ready supply of private management services in these areas from multinational organizations with their own production and processing facilities.

Box 5.1 Management contracts

The World Bank (1995) define a management contract as an agreement between a government and a private party to operate the firm for a fee (usually performance related but it may be fixed). The government remains responsible for fixed investments and holds majority ownership, ceding control only via an explicit contract, although the private contractor may provide working capital and/or may own a minority equity share. The management contract contrasts with the lease where the lessee pays a fixed rent.

Information, rewards and penalties and commitment

The World Bank (1995) use three criteria, namely information, rewards and penalties, and commitment in analyzing whether management contracts are likely to succeed or fail. They consider that competition plays an important role in increasing the availability of information, both in terms of a transparent bidding process for the management and the presence of competitive forces in output and input markets. However, the Bank argues that competition only reveals information; this needs to be used to shape rewards and enhance commitment if it is to improve performance. As might be predicted from principal-agent theory, in the World Bank survey profit-based fees or equity stakes were associated with successful contracts (improved productivity and profits) and fixed fees with no, or a relatively small, performance related component with failure. Another factor linked with success was managerial autonomy and the power to take decisions such as setting wage levels and hiring and firing workers.

Commitment from both the contractor and the government is also a factor affecting success. As with leasing, longer-term contracts with the possibility of renewal generate greater commitment from the contractor. The World Bank (1995) also state that for multinationals involved in substantial contracting work, maintaining their reputation is a strong stimulus to commitment. Governments, for their part, can signal their commitment by investing in the costs of finding a contractor and engaging in detailed negotiations, making new investments, taking politically sensitive steps such as retrenching workers or giving partial equity to the contractor (World Bank, 1995). It was noticeable that in most of the least successful contracts government commitment was reduced when donor agencies assumed many of the costs and risks associated with the contract.

Management contracts

Although management contracts can be successful, there are limits to their applicability. The World Bank (1995) consider that they are likely to work most effectively in sectors where technology is not changing rapidly, the output is a single homogeneous product such as sugar, or the supplier has an international reputation to protect. Whether management contracts are preferable to outright sale of assets obviously involves a blend of economic and political factors that will differ in varying circumstances.

Safety nets

One important area that has been neglected in many reform programmes relates to the provision of safety nets for the losers from the reform process. For example, in terms of food security there may be groups of low-income consumers who are seriously disadvantaged by the loss of food subsidies. Small farmers, particularly those in remote areas, may also find that they can no longer access marketing channels for inputs or outputs. There is also the problem that retrenchment of parastatal workers can cause considerable hardship for the individuals and households involved. This is an area that has been sadly neglected in many reform processes, particularly where the emphasis has been placed on reducing the fiscal deficit or where it has been assumed that the private sector will rapidly fill the vacuum left by parastatal withdrawal. Issues that need to be considered are whether safety nets should be provided, be permanent or temporary, and the form they should take.

5.2 How should the reforms be phased?

5.2.1 Government capacity

Pace of reforms

Decisions also have to be made regarding the pace of reforms. One school of thought would argue for a steady pace to allow a smooth transition. One factor to consider here is the capacity of the government to manage the reform process. As we have seen, decentralization either within the public sector or to the private and voluntary sectors raises a considerable number of issues. There are technical, administrative and legal matters to resolve. In particular, preparation of assets for divestiture can be a very management-intensive process. There are also political concerns. There is the need to build up a political constituency in favour of reform, to reduce opposition to the reforms and to act at an appropriate time. A second factor to consider is the strength of the private or voluntary sector and the steps that can be taken to improve it.

Another school of thought would argue for rapid reforms as a slow pace gives vested interests more opportunities to delay or derail it. Typically, this resistance is likely to be most visible when there is a lack of political commitment to reform and attempts by donors to rush the reform process are likely to exacerbate the situation.

5.2.2 New competitive pressures

Nature, extent and phasing of competition

We have argued that, wherever feasible, part of the decentralization reform process should be to strengthen the competitive framework within which agricultural services are delivered. The nature, extent and phasing of new competition can vary according to circumstances. One argument is that all barriers to entry should be removed at once to expose public sector providers to the full force of competition. There may be circumstances where such an approach may be justified, particularly where the market infrastructure is well developed or where the public sector organization has been extremely ineffective. But there are other situations, for example where the government requires a parastatal to cross-subsidize the provision of public goods from its commercial operations, where a parastatal would not be competing on a level playing field with unfettered competition. This may call for a delay in introducing competition in those functions until a hard budget had been introduced that made the cross-subsidization transparent and an appropriate subvention was being paid to the parastatal for performing this function.

Even with the removal of barriers to entry, competition may not appear immediately. For example, with parastatals engaged in marketing activities, reducing barriers to entry may lead to substantial competition in retail functions, but entry may be much slower into other functions where there are economies of size or high degrees of risk, such as wholesaling and inter-seasonal storage.

Another form of phasing is where barriers to entry are initially only lowered for certain ethnic groups or nationals of the country. One reason for this is to give a ‘first mover advantage’ to groups in society that have been excluded from, or who have not fully participated in, business activities in the past. However, if entry is tightly restricted it can also be a way of allowing favoured groups to share economic rents previously enjoyed by the parastatal. At later stages or in other contexts foreign organizations, perhaps including multinational companies might be allowed to compete.

5.2.3 Financing issues

Who should pay?

As mentioned earlier, financing issues also need to be considered in deciding on the pace and nature of the reform process. There are three aspects to this. First, there are the issues relating to ‘who should pay’, the level and phasing of cost recovery procedures and the development of suitable contractual arrangements and/or IGFTs.

Strenghtening the financial sector

Second, when decentralization involves encouraging private sector providers to compete with public sector providers, the ability of the private sector to compete or to assume functions previously provided by the parastatal, is going to be markedly affected by their access to finance for fixed and working capital. This is why financial reforms and strengthening of the financial sector are of such importance to the success of market liberalization programmes and why the neglect of financing issues may help to explain the poor or mediocre performance of many previous reform efforts. The government may find it desirable to adjust the pace, or structure, of reform to the strength of the financial sector.

Government’s divestiture choices

Third, if reform involves divesting assets, lack of domestic finance and banking facilities may make it difficult for nationals to purchase them on terms that reflect the assets’ potential earning power. This then restricts the government’s divestiture choices to:

Phased approach to reform

The presumption in this Sourcebook is in favour of adopting a phased approach to reform wherever possible, the pace being influenced both by the ability of the government to manage it and of local government and/or the private and voluntary sectors to take up functions previously undertaken by the parastatal. It has been argued already that reform is unlikely to succeed unless there is political commitment to it. Once this is attained, pressure can be put on a parastatal to improve its performance by introducing hard budgets, liberalizing trade, and reducing barriers to entry to domestic competition while ensuring that a level playing field exists. Encouraging private and voluntary sector participation in the provision of private services at market prices is likely to be much easier to achieve than the provision of public or merit goods. As government and/or parastatal regulatory expertise powers improve, private and voluntary sector organizations can be funded to produce services on contract as alternatives to parastatal production. This in turn will generate more information to improve the regulatory process. However, parastatal involvement in the provision of certain goods and services, for example agricultural research, may still be the most cost-effective alternative available even when the government’s regulatory expertise improves.

This presumption in favour of a phased approach also applies to the divestiture of assets. There seems no point in delaying the sale of assets where:

But each of these assurances can take time to achieve and this may delay the optimum time of finally disposing of those assets that the government intends to sell.

5.3 Is the reform process working effectively?

Monitoring the reform process

It is surprising to find how little attention appears to be paid to monitoring the reform process once it has been implemented. There is, however, a learning curve involved in most reform processes and providing a feedback mechanism allows mistakes to be remedied and the lessons incorporated in the next phase of the reform process. Typically the commission, ministry, committee or unit responsible for administering the reform process will not be equipped to collect and analyze the types of information required for this feedback process. Given the importance that donors place on parastatal reform it is perhaps surprising that they have not provided the funds to ensure that effective monitoring of the process occurs. This could be done either within the unit responsible for reforms or by being contracted out to research organizations and then fed back at regular intervals to the policy makers.

5.4 Conclusions

Once the decision has been taken that decentralization of the provision of agricultural support service is required, a range of political, economic and administrative factors need to be considered in deciding the appropriate form it should take. Four major general criteria are relevant in framing the questions that need to be answered in making these decisions are legitimacy, accountability, competence and financial sustainability. A general checklist of questions can be devised for each form of decentralization but this will require modifications and additions for particular situations.

From a legal and administrative viewpoint, divestiture of assets can be an extremely complex procedure and can absorb considerable managerial time. Choosing the appropriate form of divestiture, the range of potential purchasers, the method of sale and of payment all need careful consideration. Contracting out management to the private or voluntary sector can be used as a way to improve parastatal performance and to prepare it for full privatization.

Given the complexity of the reform process, the presumption in this Sourcebook is in favour of adopting a phased approach to reform wherever possible. It is important to monitor the reform process so that adjustments can be made and lessons learned from the experience.

[50] This Chapter relies heavily on material previously published in Smith (1999).

Previous Page Top of Page Next Page