Journal Issue

Reforming Transport Policy

International Monetary Fund. External Relations Dept.
Published Date:
January 1989
  • ShareShare
Show Summary Details

Lessons from Bank experience and a framework for implementing transport policy reform

The policy components of Bank operations have received increasing attention in recent years. More and more lending operations have policy reforms as their main purpose and many projects are now being specifically designed to facilitate policy reform.

Annual Bank/IDA lending for transport (including urban transport) increased from $660 million for 30 projects in fiscal year 1970 to $2,832 million for 38 projects in FY 1988. Transport lending accounted for 29 percent of the Bank’s total lending in 1970, and 14 percent in 1988. Of FY 1988 transport lending, 49 percent was for highways, 29 percent for railways, 9 percent for ports, and 6 percent for urban transport. In addition, more than half the structural adjustment loans approved during FY 1988 contained transport components, with the reform of transport enterprises and rehabilitation of transport infrastructure featuring prominently in the reform agenda.

In the transport sector, the scope of the Bank’s policy dialogue with borrowers began to widen during the early 1970s as the Bank moved toward lending to support sector investment programs rather than discrete investment projects. Typical subjects for policy dialogue at this time included: preparation of economically justified sub-sectoral investment programs, revision of tariff structures, closure of uneconomic services (particularly in railways), improving the capacity of borrowers to maintain transport infrastructure, and strengthening the performance of institutions.

More recently, following the introduction of adjustment lending at the macroeconomic and sectoral level, the scope of the Bank’s policy dialogue has widened further; policy reform components are now more explicit and are increasingly promoting reforms while strengthening domestic resource mobilization. Recent examples of the sort of policy reforms promoted under such operations include: revising petroleum product pricing, improving pricing and tax policies to support increased allocations for maintenance (especially road maintenance), and generally improving the financial performance of public transport enterprises.

How effective has the Bank been at promoting these transport policy reforms? What lessons can be learned to guide the design of future policy reforms? This article tries to answer these questions by drawing on Bank experience at large, and particularly on case studies of Bank transport projects approved between 1964 and 1982 in Colombia, Ethiopia, Indonesia, Republic of Korea, Mali, Mexico, Pakistan, Senegal, Turkey, and Yugoslavia.

Past experience

Highways. The Bank’s early policy dialogue in the highway sub-sector concentrated on strengthening operational policies to ensure that projects were implemented promptly and operated efficiently. The Bank and the borrower generally agreed quickly on the technical aspects required for successful project implementation; these did not raise political and organizational issues and did not depend on reaching agreement with other agencies.

Policy dialogue on local revenue mobilization—to ensure the government could provide adequate counterpart funds to complete the project and maintain it to a reasonable standard—had a mixed outcome. Governments generally provided sufficient counterpart funds to complete projects, but rarely provided enough for road maintenance. Various factors made allocation of sufficient maintenance funds difficult: (1) increased allocations could generally only be discussed in the context of the government’s overall fiscal balance, and this was rarely done; (2) reallocation of funds from the capital to the recurrent budget touched on sensitive political issues (construction frequently has higher political priority than maintenance); and (3) governments were not convinced that the benefits of economic efficiency pricing, which would have raised more revenue for maintenance by increasing the tax on heavy vehicles, were sufficient to offset the objections of the trucking lobby. On the other hand, borrowers were receptive to the Bank’s request to raise fuel prices to international levels, so as to promote energy conservation, raise revenue, and reduce smuggling to neighboring countries; widespread public knowledge of the energy crisis made it easier for governments to raise these prices.

In general, the Bank was successful in strengthening the technical capacity of borrowers, through provision of technical assistance and training over a series of operations which acted to reinforce each other over a period of 10–20 years. The Bank’s success in this area seemed to be attributable to the uncontroversial nature of the required actions, the absence of a need to consult and seek agreement from other agencies, and the evident benefits (more Bank-financed projects) which such technical improvements produced. However, where the Bank sought to combine such technical improvements with recommendations for reorganizing the executing agency, it encountered much more difficulty.

Attempts by the Bank to persuade borrowers to enforce existing regulations on vehicle dimensions and axle loads were rarely successful. Existing regulations were often inadequate and new legislation was needed to make them effective. Enforcement was also hampered by the fear that it might significantly increase freight rates and that greater enforcement efforts might encourage corruption. Governments also recognized that, since the trucking lobby was a powerful pressure group, the enforcement of axle weight regulations would require the provision of some sort of compensation.

Railways. In most railway projects the Bank attempted to improve costing and accounting procedures and to raise tariffs to more economic levels. It generally succeeded in achieving the first two goals, but not the third. The public invariably resisted increases in passenger fares, arguing that they should not be expected to pay for the operational inefficiencies of the railways, while government departments (usually significant users of freight services) vigorously opposed raising freight rates. Moreover, the frequent need to seek formal government approval for tariff increases made changes cumbersome and time consuming.

In several cases the Bank requested that uneconomic railway lines be closed to improve the financial health of the railways and, where low-cost transport was deemed desirable on social grounds, that such services should be provided using cheaper alternatives. Railway lines have an important symbolic value and decisions on closure must frequently be referred to the highest levels of government, where political costs are often found to outweigh potential economic benefits. The closures that were successful proceeded slowly and usually depended mainly on the government’s political strength and the manner in which they were presented to the public. In some countries, closures were facilitated by building a low-standard road alongside the line to be closed.

In all the railway projects reviewed, the Bank attempted to strengthen the autonomy of railway management to ensure that financial and operational decisions would be based on commercial rather than socio-political criteria. Implementation was difficult and slow, since greater emphasis on commercial decision making nearly always conflicted with the inherently political nature of decisions about closure of uneconomic lines, suspension of loss-making passenger services, raising tariffs, and displacing surplus workers. The attempts also had implications for legislation and human resource management; it takes a long time to draft, agree on, and pass the legislation needed to make railways more autonomous and it also takes time to create a commercial outlook in staff with a traditional public-service orientation. Progress was minimal, except where the initiative for change came directly from the government or railway management, or where the Bank took the drastic step of suspending disbursements. More recent experience with “contract plans” and other approaches to improving the relations between government and transport enterprises—so as to clarify their financial goals—appears to have been more successful.

Ports. As in its railway lending, the Bank frequently attempted to encourage greater autonomy for ports, with a view to strengthening financial management, enhancing the authority of port managers and providing control over a portion of the port’s foreign exchange earnings to facilitate prompt procurement of fuel and spare parts. These attempts generally failed to achieve their stated objectives; though governments did agree that management improvements were needed, they did not believe that such improvements could only be achieved through greater port autonomy. Also, greater autonomy threatened too many vested interests. The concept raised questions about the management of other public enterprises and the proposed retention of foreign exchange earnings ran counter to established government procedures. Full autonomy generally also required changes in legislation. Substantial progress was made in establishing commercial accounting systems, revaluing port assets and improving the overall financial viability of ports by increasing productivity and raising tariffs. In contrast, efforts to introduce cost-based tariffs were generally unsuccessful. Improving financial discipline and maintaining revenue levels offered clear advantages to the port authority. Cost-based tariffs, on the other hand, would have required more accounting staff and port managers did not believe that the structure of port charges was a sufficiently important factor in the decisions made by shippers.


First, while the Bank has succeeded in promoting certain types of policy reform, it has been less successful in reforming policies which involve significant political and organizational issues. This lack of success appears to be attributable to the Bank’s tendency in its analytical work to focus exclusively on the technical, economic, and operational aspects of major policy reforms, without recognizing their inherently political and organizational nature.

Second, policy reforms promoted by the Bank tended to be more effective when they dealt with technical issues where the necessary decisions could be taken by a single agency. When more than one agency was involved, the reforms encountered considerable resistance from the other agencies and were generally unsuccessful within the time frame set for them. This was partly because the Bank had the right audience for technical discussions—the executing agency—but the wrong audience for discussions on institutional matters. For example, decisions relating to greater autonomy for public transport enterprises are generally taken by the highest levels of government; given the breadth of their concerns, people at this level are less readily influenced by the technical arguments of Bank transport staff. The reforms might have been more successful had they been supplemented by the dialogue conducted under Bank-supported structural adjustment and public enterprise reform operations.

Time required for typical policy reforms(years)
ActivityDeregulating transport industriesRevising structure of road user taxesClosing uneconomic railway linesPrivatization
Preliminary agreement on the need for the reforms 13–52–32–33–5
Detailed examination of the options available and the likely consequences of the reforms1–22–31–21–2
Preliminary consultations with parties likely to be affected by the reforms13–52–31
Attempts to avoid reforms through rationalization (or experimentation to establish likely consequences of reforms)22–33–42–3
Implementation of reforms (detailed specification of policies, establishing new laws and legislation, and revising administrative arrangements)2–32–32–32–3
Introduction of new policies (transition stage)11
Total time needed10–1510–1510–159–14
Source: Review of Bank-supported projects.

Generally involves discussions and preparation at Terms of Reference.

This stage is often omitted.

Source: Review of Bank-supported projects.

Generally involves discussions and preparation at Terms of Reference.

This stage is often omitted.

Third, the Bank’s economic and sector studies tended to be good at providing broad policy advice in the transport sector, but were often weak in identifying the origin and rationale for existing policies and developing a practical reform program tailored to the country’s circumstances. Moreover, most reform programs advocated by the Bank seriously overtaxed the borrower’s administrative capacity to process and implement the reforms, even when the reforms were fully understood and the government wished to implement them. This was particularly true in the institutionally weak countries of Sub-Saharan Africa.

Fourth, policy reforms are time-consuming and cannot easily be completed in the five to seven years typically specified in Bank operations (i.e., within the lifetime of one investment or sector project, or two consecutive structural adjustment loans). A review of Bank projects showed that deregulation of road transport usually took over ten years, while closure of uneconomic railway lines took about ten years for important lines and about four to six years for very low density lines. Experience in developed countries is similar. The time-consuming procedures involved in replacing one major policy with another generally mean that major policy reforms must be implemented over a sequence of several Bank operations, with each operation attempting to complete an agreed set of actions forming part of an agreed long-range program.

Fifth, a brief review of the transport components of recent structural adjustment loans (SALs) does not suggest that SALs, by themselves, are more effective at policy reform than regular sector operations; however, it does suggest that SALs can play an important supportive role. Nearly all the components reviewed were prepared under earlier sector operations and had already appeared as loan conditions in operations already in progress. The role of the SALs was to provide mutually reinforcing loan conditions; since these conditions were generally targeted at the ministry of finance, they ensured that transport policy issues being discussed with the Bank received more attention from the finance ministry and this helped to quicken the pace of reform.

The case studies repeatedly emphasized the need for a better analytical framework for analyzing transport policy reforms and for designing policy reform components. The last part of this article builds on these insights.

A framework for reforms

Policy reform components need to be designed and implemented with the same care and attention as the physical elements of a project. The proposed policy reforms need to be identified on the basis of a detailed country analysis, paying particular attention to the origins of present policy. The analysis should demonstrate that the proposed reforms are the simplest available, particularly when they call for major institutional changes, and that the same aim could not be achieved through better use of existing policies and institutions. Dialogue should furthermore be confined to issues which the government and the Bank agree are timely and relevant; the Bank cannot force borrowers to make policy changes which they consider to be irrelevant to the goal of the lending operation, unlikely to offer significant benefits, ill-timed, or to involve unacceptable political risks.

Within this framework, the design and analysis needs to take account of the following factors:

Benefits and costs. These need to be identified, even if they cannot be quantified, in order to show that the benefits of the reforms are likely to exceed the costs and that the solutions proposed are the best alternatives available, and to establish priorities. It is necessary to ensure that the costs identified are recognized by the government and that resources are available to meet them. For example, if the policy reforms necessitate a scheme to compensate workers made redundant, finance will be needed to meet these costs.

Administrative constraints. The design of policy reform components needs to take account of the government’s capacity to process and implement the reforms. The administrative resources likely to be needed for a proposed reform can be analyzed with reference to five steps:

  • Preliminary discussions leading up to agreement on the need for reform.

  • Examination of the options available and the likely consequences of implementing them.

  • Consultations with interested parties likely to be affected by the reforms.

  • Implementation of the reforms:

    • detailed specification of new policies

    • changes in laws and legislation

    • revision of existing administrative arrangements.

  • Introduction of the new policies, possibly over a graduated transition period.

If the administrative requirements exceed the government’s resources, a sharply focused technical assistance program will be needed, or the reform program will need to be scaled down.

Political and organizational constraints. Though the Bank is specifically prohibited by its Articles of Agreement from participating in the political affairs of its member countries, it must recognize political realities if it wishes the policy reforms it supports to have a reasonable chance of success. The designers must therefore recognize the political and organizational constraints likely to hamper implementation of the reforms-deciding which stages of the reform process can be implemented without difficulty, which will require the provision of compensation to overcome these constraints, and which involve such sensitive issues that implementation must await more favorable political conditions.

Timetable for reform. Overall, Bank experience suggests that the implementation of some typical transport policy reforms may take about ten years and will follow the broad pattern outlined in the chart. This chart shows how, by disaggregating a proposed policy reform into its component parts, one can estimate how long it will take to accomplish the reform. The length of each stage clearly depends upon the administrative, organizational, and political structure of government and can be lengthened, so as to reduce the need for additional administrative resources, or shortened, at the expense of raising its administrative, organizational and political costs.

Conditionality. In the transport sector the conditions attached to Bank loans, particularly those relating to actions to be taken before loan approval, play an essential role in encouraging policy reform. After loan approval, they play a smaller role. However, conditionality after loan approval continues to play a useful supportive role by establishing a clear timetable for the reform agenda, helping to overcome minor obstacles to reform, and serving as a record of agreements reached in order to monitor project implementation.

Other Resources Citing This Publication