Spring Meetings 2004 Dialogues

The purpose of this session was to discuss economic policy conditionality and the Action Aid Report: "Money Talks: How Aid Conditions Continue to Drive Utility Privatization in Poor Countries." John Garrison from the Civil Society team at the World Bank opened the session with a general welcome and introduction of the panelists. Patrick Watt, Political Economist who works for the Aid and Accountability Unit of Action Aid UK; Stefan Koerberle, Senior Economist in the Operational Policy and Country Services Unit of the Bank; Nancy Alexander, Director of Citizen’s Network for Essential Services and Jan Janssens, Lead Specialist from the Water Supply and Sanitation Unit at the Bank.

Meeting Notes

Patrick Watt made a presentation of the report. The three other panelists commented, and a discussion followed with two rounds of questions and comments.  Below is a summary description of the session including an overview of the report, the presentations and the ensuing discussion.

Patrick Watt: The report focuses on the use of conditionality with specific reference to the electricity and water sectors and it the deals primarily with economic policy conditionality. The report is not talking about fiduciary conditionality, or about governance or process conditions, and it is primarily a report on conditionality not privatization.

Action Aid has four key criticisms of conditionality:

• Conditionality is unfair. It reflects an asymmetrical relationship between the World Bank, the International Monetary Fund (IMF) and bilateral donors on the one hand and aid-dependent low-income countries on the other. The result of conditionality in many respects is an abuse of power.
• Conditionality is undemocratic. In skews accountability relationships. It gives technocrats from the IMF and the World Bank and other donor agencies significant policy influence without any accountability for those policies. And it makes governments accountable to the donor agencies more than to their people.
• Conditionality is ineffective in the long run. In the absence of strong domestic ownership of policy reform, conditionality will rarely lead to sustainable policy change. 
  
• Conditionality is inappropriate. Conditionality reflects a lack of confidence in the ability of government and local civil society to identify their own policy choices and it tends to lead to reform that does not actually reflect local conditions.

There have been positive changes in Bank and IMF policy in recent years and new commitments to reduce the use of conditionality. These include: the Poverty Reduction Strategy (PRS) process which uses the language of partnership and ownership and is placing increasing emphasis on the need for donors to act in a mutually accountable way in development compacts with recipient countries. It also includes new instruments like the Poverty Reduction Strategy Credit (PRSC) which is designed to operationalize the primary commitments of the PRSP. There is also the commitment in the IMF through Poverty Reduction Growth Facility (PRGF) to harmonize and coordinate effectively with the Bank. Another positive change is the recognition by the Fund that it had strayed too far into the use of structural conditions, so since 2001, a commitment from the Fund to pull back from structural conditionality. And finally there has been a commitment from the Bank to move away from the specific promotion of economic reform blueprints.

This report looks at what is happening on the ground in order to determine the extent to which these commitments are actually being operationalized. And the conclusion from  a review of the project documents is that while conditions have changed in terms of how they are reflected in Bank documents they have not actually gone away. The condition, for example, of privatization, may be less explicit but it is still there and the Bank does not appear to be supporting a more heterogeneous variety of service delivery options on the ground. [Note: these points were illustrated with three cases in the report: cases from Uganda, Ghana and India]

Other findings included:

There is wide spread use of selectivity along side conditionality to push policy preferences in the utilities. An example is the use of the Country Policy and Institutional Assessment (CPIA) which raises the risk of a lot of de facto prior actions that are being agreed to with governments as a precondition to moving toward lending agreements.

There is also an increasing use of cross-conditionality as bilateral donors move toward providing budget support and aligning their own bilateral budgets with PRSC conditions. Finally policy advice and technical assistance are playing a very powerful role.

In terms of the effect on poor people, the report looked at a number of issues and found no evidence of positive impact from the changes. Similarly, the Bank’s internal Operations Evaluations Department (OED) recently reviewed 150 Bank power projects and found that there was essentially no data to gauge poverty impact. In the case of the water sector a recent OED evaluation of the Bank’s water strategy came to similar conclusions.

In the report, Action Aid puts forward the following recommendations:

• End economic policy conditions and restrict other conditions to what is necessary to ensure that money actually gets spent on intended beneficiaries.
• Reform technical assistance. The whole way in which policy advice is being dispensed at the moment to poor countries is being used as a support to the use of conditions. It needs to be much more neutral. 
• Conduct a fully independent review of the use of conditionality by the international financial institutions and donor governments and its impact on the poor.

Stefan Koerberle:  We agree with some of the points in the report, especially the point that we want to make sure aid is spent on the intended beneficiaries and is not used to push specific conditions in low income countries. Where we would have some issues is on some of the points that are overstated on conditionality.

First of all, conditionality is seen in Bank not as a way to push certain policies or to make countries do something they wouldn’t otherwise do. Conditionality can not be a substitute for country ownership. That is one of the key conclusions from failed operations of the past. Country ownership is one of the key preconditions for any effective aid delivery. Also in terms of selectivity, it only makes sense to engage in a policy dialogue if the country is actually interested in particular issues and is actually seeking advice and is actually trying to come to a policy conclusion and trying to implement certain measures.

Some of the specific charges about conditionality: first on the undemocratic nature of conditionality. The World Bank is often accused of being undemocratic and not accountable but there are several kinds of accountability the Bank has: 1) The Bank is a technocratic institution but it is also a profoundly political institution in the sense that it does have a board and it is accountable them; client countries sit on that board, and board members are accountable to their governments. 2) The Bank is not doing things in a particularly secretive way. Most of the documents we have are disclosed and 3) Bank staff are quite happy to engage in open discussion.

Where we do have an issue is in participation. Participation is a key ingredient of success in many of these policy-based operations, but the kind of participation varies from country to country. It can’t be the same in every country. It is difficult for us to say on the one hand we recognize country ownership and on the other hand push countries to adopt certain procedures in terms of participation. One point that is often overlooked both in terms of accountability and participation is that the World Bank engages through loans that are essentially international treaties and each one of those has to be approved by parliament in a country.

On the unfairness issue of conditionality, you highlighted the point that it is abuse of power. I have two observations. One is when you are not inside the Bank, you fall into the fallacy of overstating the power and influence of the World Bank. The second is conditionality is actually a two way street. Its a mutual commitment device. It actually engages the country in a more structured way and allows the country to spell out the road map by which it achieves the results that it has actually set out in its PRSP or in its overall poverty reduction strategy. The role of the World Bank or other donors is to help the country articulate and operationalize that strategy and vision.

On the effectiveness/ineffectiveness of conditionality; we agree that especially in the past in many cases conditionality has been ineffective. Its been particularly ineffective when conditionality was used as a substitute for government commitment or when it was used as part of policy-based lending for countries that were essentially political cases. But on the point of ineffectiveness, it is an overstatement to say that conditionality on average tends to be ineffective and that conditionality is a powerful tool that pushes countries to do things they wouldn’t otherwise do.

Finally, policy advice is a core part of what the World Bank does and we don’t actually want to see the financial support disassociated from policy advice.

Nancy Alexander: I would like to provide a larger context for looking at conditionality around utilities by identifying three dynamics that drive governments to implement donor and creditor policy preferences, even if those preferences clash with those produced through open, democratic processes.

1.  Conditions for Eligibility to Aid/Credit.  Traditional types of conditionality make credit contingent upon the promises of governments to execute specified policy conditions.  In addition to traditional types of conditionality, there is a trend toward a priori eligibility conditions that must be met in order to access aid.  Many cash-strapped governments look to external assistance as their lifeline.  A priori eligibility conditions put them under heavy pressure to conform with donor and creditor policy preferences.  This pressure may override the policy preferences produced through open, democratic processes.  Harvard’s Dani Rodrik says that governments should be able to “opt out” of conditions that pressure governments to flout the will of their people.

The Bank rates the policy and institutional performance of all borrowing governments with the “Country Policy and Institutional Assessment” (CPIA) and, for low-income countries, this rating is a significant factor determining their access to IDA credits.  The World Bank practices “selectivity” when it allocates credits to countries with, among other things, willingness to conform to its policy prescriptions.  The CPIA also determines the policy thrust in a country so if a country is weak in terms of say trade liberalization or privatization, then the World Bank’s lending strategy for the country – its Country Assistance Strategy (CAS) -- will focus on remedying that “weakness.”  The Bank’s “CAS Retrospective” (March 2003) describes how CAS increasingly focus on remedying CPIA weakness and draw from the PRSP only as they reinforce policy preferences embedded in the CPIA.

The same type of rating determines access to credit at a subnational level.  Someone said that the Bank is a very small player in borrowing countries, such as India.  However,  India has 24 states and most Bank aid goes to three states that are implementing an agenda that they crafted with the Bank to undertake some significant liberalization and privatization, so 21 states are receiving a good deal less money, and sometimes the poorer ones. 

2.  Decentralization and regionalism.  Applying conditionality to subnational – state and local – governments is much easier as the IMF and World Bank promote decentralization.  It is also easier as the multilateral lenders extend credit for regional processes where democratic engagement may be more indirect.  

In terms of decentralization: responsibility for infrastructure – electricity and water – and social programs are being decentralized to provincial, state or local governments.  In order to gain access to international capital markets, subnational governments are required to raise a significant and increasing share of their own revenue, rather than rely so heavily on transfers from the central government.  This dynamic reinforces conditionality that promotes commercialization and private sector participation in electricity and water infrastructure. 

Decentralization usually creates macroeconomic instability and, in response, the IMF pressure subnational governments to raise taxes and fees for basic services.  As subnational governments are left with “unfunded mandates,” there is pressure to privatize. When this fiscal pressure results in “private provision by default,” democratic preferences and citizen voice become irrelevant.  Particularly in the area of essential services, the result of marginalizing citizens can be dire.

In terms of regionalism we are seeing, for example, about 10 regions in the regional infrastructure project in South America (IIRSA) and additional infrastructure regions laid out in Central American in Puebla Panama. In a regional system, it is more difficult for infrastructure-related decisions to involve and be accountable to citizens of the countries involved...

3.  Loan instruments that pressure recipients to introduce private provision of electricity and water.  (The IMF, WB and WTO define the “private sector” as comprising both private firms and not-for-profit organizations.)  A) The IMF/WB can pressure governments to cut off subsidies to electricity and water utilities and break up cross-subsidies among sectors (telecommunications to water).  The resulting fiscal pressure can force privatization “by default.”  B)  Donors, creditors and governments can “pool” their resources and only extend them for the purpose of introducing private provision of utility services.  C)  Certain types of loans (adaptable program loans and community-drive development loans) are primarily available to those governments (including subnational governments) willing to adopt private provision.  For instance, the Bank is scaling up community-driven development (CDD) programs in rural areas in particular and in terms of water services, we really need to look at CDD programs because of the scale. In Africa one half of IDA’s money is going to scaling up CDD programs and it is important to realize that these programs are contracting out water, health care, education – often with significant quality problems.  For instance, in terms of water, only 24 percent of the water components of CDD programs  are sustainable.

In summary, to claim or reclaim open and democratic decision-making processes, it is important to not only understand policy conditionality, but also the dynamics created by donors and creditors that put governments under significant pressure to adopt their policy preferences.  Sometimes open and democratic processes support the policy preferences of donors and creditors and oftentimes they do not.

Jan Janssens:  I want to make four points:

• First, I would submit that there are better sectors to illustrate the arguments in the report than water. Water is a very special sector because it includes  infrastructure, private-sector development, health  and so on  It presents a complex situation, and we have to bear that in mind.
  
• Second, I am convinced that  we all in this room have the same basic objectives, namely to have better quality service which is safe, reliable, with access to all.
  
• Third, providing physical investments is  not really sufficient, providing the right enabling environment is equally important. In the 1980s, during the “water decade”, many mistakes were made because the two were not connected.  Most of the investments in the sector that were made in the early part of the decade were not or insufficiently maintained. After five years these investments were in bad shape and needed urgent rehabilitation. In many cases donors were asked to come back to these countries to refinance what they had financed five years before. An additional problem was that inappropriate technology was often used which was hard to maintain  A policy reform process, when needed, should go hand in hand and very often upstream of implementation of the physical investments.
  
• Finally, on private-sector participation, my own experience leads me to believe that this has never been a conditionality in the terms of an economic conditionality. We are trying to work with all of those involved in the sector in getting sustainable accountability and autonomy of service providers. This is not an easy task and it certainly varies in different countries. There is no blueprint for this and thus it is always a challenge. In a number of cases a private partner is a kind of insurance policy to guarantee autonomy if there is a change in the political situation.  But remember still 85 percent at least of the world population is still being served by the public sector, and we have to work to strengthen those systems where needed.

top

First Round of Questions /Comments

Q -- The first participant challenged the Bank and the IMF saying: You are talking about good intention, but in reality, I think there are different things happening on the ground. Although the Bank promises that they are not pushing the country to adopt certain conditionalities, it seems otherwise and its seems that as a result of their macroeconomic policies there is an increasing the gap between the poor and the rich.

Q -- The second participant from Action Aid Uganda had several points. First, he said that Jan Janssens had said that privatization is not a conditionality but this is a measure included in the Uganda PRSP. Second, in Uganda, where there used to be subsidies for low-income consumers, now they are paying more and industry is being subsidized instead. Third, in the Uganda water sector a private company was hired to manage the water supply but the company externalized its profits.  Before this the national water corporation was profitable and generating lot of income.  Fourth, in terms of the inappropriate technology  examples cited, it is important to know  who exactly financed these projects.

Q -- The third participant wanted to know if the Bank was working with public utilities in Kinshasa and other places.

Jan Janssens: On the issue of the type of technology employed, the example I mentioned was not funded by the World Bank, it was financed by another donor agency, but it was clear that it was inappropriate technology.  In these situations, one can blame the contractor the donor, but sometimes you also have to blame the recipients. Second, on the conditionality of the PRSP in Uganda, it depends on how you interpret the text, as these are often the result of a long process of discussion between the Bank, government officials, and other stakeholders. On subsidization, the whole subsidy discussion is complex and there are a often errors of inclusion and exclusion.  There is active work going on to document and assess experience with subsidies in both the water and energy sectors using data from a number of good practice case studies.  A hardcopy publication of this work will likely be available by end of FY05.

Patrick Watt: It is important to be very clear about what we really mean when we talk about ownership because there are some very different versions or understandings of ownership, and that actually lies at the heart of a lot of disagreements on conditionality. The Bank is very clear about what ownership entails.  Ownership is a means to an end. Ownership is defined as the likelihood of a reform being carried out, it is perfectly possible to have ownership under that definition and for the reforms to have been identified developed and approved outside by external agents. Ownership very rarely extends beyond a small technocratic group of donor staff and senior government technocrats, and that underlies a lot of the problems and controversies that all of these privatization projects have generated. These arguments should to be carried out in a kind of marketplace of ideas in a way that is not happening at the moment.

Regarding the issues of water conditionality I think it is simply untrue to say that these conditions aren’t out there, now to some extent those conditions may be a result of policy dialogue that has taken place between the Bank and the government, but they are out there.

On the issue of  power and the Bank’s agenda,  we are very clear in the report that we are not saying that the Bank says jump and the country jumps. There is a dynamic process in negotiation. The point is that in many low-income and aid-dependent countries, the process of negotiations is skewed from the outset. That is our argument. Fundamentally, this is a discussion about what level of policy influence is appropriate for external agents to exercise and at the moment it is clear that the degree of policy influence exercised by the Bank and Fund is wholly inappropriate.

Regarding the argument that there is an apparent contradiction between saying conditionality is ineffective on the one hand and intrusive on the other, I don’t think there is at all. Even if conditionality doesn’t directly translate into the results that the Bank and Fund are looking for, it distorts the accountability relationships and it actually diverts a lot of political energy away from enabling countries to identify their own policy priorities.

Nancy Alexander: The World Development Report (2004) has a very frank assessment of the huge power that donors and creditors have collectively and the leadership that can be exercised. It talks about how donors and creditors should use their influence and it talks about how donors and creditors wield 20 percent of the development finance in about 60 countries and over 40 percent of development finance in 30 countries. What we are seeing now is a shifting paradigm. Conditionality was front and center, now there are ex-ante mechanisms like the CPIA that exert huge pressure over a country’s policy choices including around utilities. These involve conditionality indirectly not directly.

top

Second Round of Questions /Comments

Q -- The first participant wanted Nancy to clarify her statement about decentralization leading to macroeconomic instability. He also wondered if the impact of decentralization had been studied since  one would assume that if countries are shifting from central government financed programs to decentralized local service provision, this would lead to the kind of regional inequalities experienced in the US where poorer regions can’t have the same quality of schools as the richer communities.

Q -- The second participant raised a point about the lack of private investment in infrastructure especially for poor areas for profitability reasons, and referred to a paper from August 2003, called Public Money for Private Infrastructure about developing instruments and deciding when to offer guarantees to address this problem This paper he said was about making it profitable for inherently unprofitable private operations to go forward.  He also read from an operational document at the Bank titled Public and Private Sector roles in the supply of Electricity which seemed to suggest using Bank instruments for this purpose. 

Q -- The third participant who was interested in the water sector brought up a report, by Michel Camdesus about financing infrastructure in the future that, she said, pointed out that the Bank has to look more broadly at public and private opportunities, and not solely focus on privatization. But, she said, looking both at what’s in the pipeline and looking at projects over the past five years it is clear that privatization is involved in about 90 percent of the cases or more. So is there really space for this broader perspective or is the World Bank only interested when the government chooses to privatize?

Stefan Koerberle: There are a variety of approaches that the World Bank is supporting,  Vietnam, for example, where state-owned enterprises are delivering even commercial services. Vietnam has actually been a very successful borrower from the World Bank and has reduced poverty quite significantly. So you do not see conditionality on privatizing there because clearly the country has said that is not what they want to do.

On the CPIA, it is one mechanism and a fairly broad one at that.  Its actually not unlike a scorecard, it focuses on things that the World Bank generally accepts as part of a good policy environment. It reflects the Bank’s thinking that lending is only successful, or only leads to good development results, in a policy environment that is broadly perceived as conducive. There is no point in lending in an environment that is not conducive to development effectiveness. That is one of the lessons we have learned, so the CPIA tries to capture that and aid allocation tends to be larger for those countries where we think there is a conducive or enabling policy environment.

Jan Janssens: The Bank is generally engaged with a full range of public and private companies.  A study on modes of engagement with the public sector in water supply and sanitation is being carried out.  A hardcopy of the report will be available by the end of FY05.

On the cited operational document, one should be aware that public finance alone is not going to be sufficient.  We have to find ways on how to bring together public and private, in so called hybrid financing schemes for water. The public and the private sector need each other. The challenge is how to bring private finance in to what has been traditionally a public sector and how to have the public sector come into a private-sector environment with the right incentives.

Nancy Alexander: I can give you many, many sources that say that macroeconomic destabilization is a necessary component of decentralization because it expands the overall spending.

Patrick Watt: What is needed is Bank support for a much more systematic and independent evaluation of the use of policy conditionality, because there is always a gap between policy and practice. That would be a good way of taking us forward.

John Garrison: In closing, there is clearly a lot of interest in this subject.  I hope that participants felt that this was a useful debate.   I think it was despite the fact that there is still a good deal of tension and controversy in the room which, of course, is also part of the process.  I agree with Patrick that we have different viewpoints and perspectives not only on adjustment policy, but on the whole issue of country ownership and democracy.   Until we understand each other’s positions better it is really going to be hard to agree on a common set of data and analysis. But I welcome further debate.  Finally, we would be happy to try to set up separate meetings between some of the participants and Bank staff on Uganda as well as on the issue of conditionalities in the electric sector.

top

Back to  2004 Spring Meetings Dialogues with CSOs page