Ethiopia and the World Bank in the 2000s

The role of donors in Ethiopia has emerged as a focus of an ongoing debate on authoritarian aid. (See here and here) As part of a broader wariness of where  moralizing hubris  can lead, I made a decision a few years ago to focus my work on the challenges of ‘messy democracies’ (which, evidently as of 2021, includes the USA and other Northern countries). Though I don’t intend to engage with the current controversy, between 1999 and 2008 I was quite centrally involved with the World Bank’s engagement in Ethiopia, especially from a governance perspective; I subsequently  wrote about that experience. Here, as a contribution to the current debate, is the relevant extract from my 2014 book, Working with the Grain.

“Ethiopia – planting seeds of bottom-up accountability.  Between 2000 and 2010, aid inflows averaged about 5-8 percent of total annual income (in the range of $1 billion annually) – and accounted for about one-third of public expenditures. When it comes to aid, mutuality often  plays out in a troublingly superficial way: political support from citizens of donor countries depends importantly on the aid effort’s ability to evoke among ‘Northern’ taxpayers a warm feeling of doing the right thing.

Prior to Ethiopia’s 2005 election, the country had become a poster-child of ‘good’ aid. Back in the 1970s, along with Bangladesh it had been the country where images of starving children had evoked a rash of ‘live aid’ rock concerts and feel-good donations. For a while, the brutality of the  repressive military Derg regime undercut the narrative. But finally, with the emergence in Meles Zenawi of  a new-generation-leader committed to development, the narrative could come together.  The strength of   commitment by donors to Meles Zenawi’s government was evident both in the amount of aid, and in the form in which it was given. Ethoiopia became a leading example of new, cutting edge approaches to development aid. 

A common criticism of aid is that it supports gold-plated enclaves (complete with the donor country nameplate)  in the form of  initiatives which destroy the capacity of national governments by undercutting the recipient government’s ability and willingness to make choices, and by  luring the most talented people away from the public sector. In response to this criticism, in countries where governments seemed committed and capable, donors increasingly were moving to provide aid as annual ‘budget support’ for the country’s expressed priorities. (This isn’t quite the blank check it seems. It provides a platform for in-depth dialogue between donors and recipient governments as to priorities and performance. As champions of budget support pointed out, having some influence over all of government spending was surely likely to do more to combat poverty than having direct control over what rarely amounted to more than 5-10 percent of the total spend.) Meles’s commitment to development, plus the country’s track record of managing resources prudently, had made Ethiopia a major recipient of budget support.

But in the violent aftermath of the 2005 election, the positive story came undone. It became politically impossible to write an annual budget support check; that would signal seemingly unqualified support for the Meles regime. Instead, the clamor arose for donors to withdraw support entirely from Ethiopia. What was to be done? Donors adopted a two-part response.

One part was a fig-leaf of sorts. In place of budget support, and without cutting aggregate levels, donors embraced a new aid model for Ethiopia: the protection of basic services. Formally, there were two large differences between the old and new models. Aid no longer was made available for general purposes: it was specifically targeted to support a scaling-up of social sectors by paying the costs of teachers and health workers. Better yet, in Ethiopia’s radically devolved formal constitutional arrangements,  education and health were the functions of regional governments, the support provided was no longer going directly to Meles. In practice, though, budget revenues that aren’t used for one thing can be used for another. Provincial levels had no independent revenue-raising capabilities, and teachers and health workers were already being paid indirectly by the center through inter-governmental transfers. But budget fungibility is an argument  for technicians. Viewed through a more political lens, the advantages are large vis-à-vis donor country electorates of reframing aid  in terms of direct support for teachers, nurses and doctors.  

The second part of the donor response also might initially have seemed symbolic – though it was especially difficult to negotiate with the Ethiopian government. In return for large-scale continuing aid support for the provision of basic services, donors pressed hard for the introduction of a variety of bottom-up mechanisms to enable citizens and civil society organizations to monitor whether public resources indeed were delivering on their intended purposes.

Implementation was a long, slow process; for four years, there were repeated disagreements between donors and government, and associated delays. But, remarkably, the Ethiopian authorities themselves increasingly have embraced the bottom-up approach. As of 2012, over 3,000 officials from across the country had been trained in how to design and implement good practices in local-level financial transparency and accountability; over 50,000 local leaders have been sensitized as to how they can proactively monitor public spending; over 90 percent of all local governments were posting budgets.

To be sure,  no one would confuse contemporary Ethiopia with a vibrant, multi-party democracy along the lines of contemporary Korea.  Meles’ regime was not one to make the same mistake twice. Going into elections in 2010, there was little doubt as to the outcome. In the event,  the EPRDF won close to two-thirds of the vote, and 99% of the seats in national and regional parliaments. But the journey of development along the dominant trajectory can be a long and surprising one. In the early 1960s, no one would have predicted that forty years later Korea would be a thriving multi-party democracy. Whether Ethiopia can sustain a further two decades of stability and broad-based, inclusive economic growth is  enormously uncertain – and Meles’ untimely death only underscores the risks. But if Ethiopia is able to remain on its current trajectory, the seeds of better governance which have been planted over the past two decades – a de jure democratic constitution with strong formal checks and balances; and a de facto willingness to explore how bottom-up transparency can help hold public officials accountable for performance – could yet be early harbingers of a profoundly transformed polity and society.” (2014; pp. 65-67)

Anti-Corruption for Adults

A new publication  from the World Bank offers a view of the past, present and future of work on anti-corruption. The vista left me feeling surprisingly hopeful.

‘Anti-corruption’ never has been my favorite topic; its discourse has, to my ears, often seemed  too self-righteous, potentially counterproductive in its polarizing, Manichean certainties.  But Enhancing Government Effectiveness and Transparency: The Fight Against Corruption, published by the Bank in September 2020 is different.

From the start, unlike  many anti-corruption dicta (some of which, as a one-time insider, I helped to write) the report foreswears any grand statement of what is to be done. Rather, it frames its purpose as:

“ to equip public sector officials and civil society with a set of approaches, entry points and tools that can be drawn upon and adapted to their specific country context….. By design, many of the examples are taken from countries and sectors where corruption remains widespread but where public officials and civil society have not given up battling it.”

There is a sense of sobriety, of sharing hard-won insights.  No quick fixes. No ‘best practice’ technocratic solutions ready to be parachuted in. There is recognition of the centrality of context:  

“Political realities  constrain the menu of policy responses from which leaders may choose…… Technical solutions alone are insufficient to have an impact on corruption, nor can they be merely transplanted from one country context to another”

 “Many high-profile anti-corruption strategies have proven to be ineffective and only give a veneer of government action…..Progress is not linear and reforms could suffer due to political setbacks and/or institutional weaknesses, yet even basic efforts could provide a foundation on which to build.”

‘Magic bullets’ from times past are given no special treatment:

“Pervasive institutional limitations raise questions as to whether the model of a stand-alone multi-functional anti-corruption agency is the right one…..”

“While many countries have asset and interest declaration systems, there is limited evidence of their effectiveness. Most AID systems have yet to live up to their potential.”


“Progress in the fight against corruption is not necessarily from the large government-wide announcements and initiatives that garner extensive press coverage, but from the more focused efforts and more subtle changes that governments and communities make that may go unobserved.”

Having thus cleared the air, the report directs attention to practical entry points for change, organized around  brief overviews and in-depth case studies for each of fifteen anti-corruption-related topics. Even though the authors explicitly foreswear the ambition to offer  a grand synthesis, there seemed (at least to this reader) to be an implicit  unifying thread – the idea that entry points emerge and expand  through creative interaction across multiple dimensions, with pride of place given to interactions  among sector-level initiatives, stakeholder engagement, and ‘govtech’.

On sector-level interventions:

“Unpacking sector-specific issues is crucial to diagnose the root causes of corruption in public services and design appropriate interventions. Even within a sector, a reform could be narrowly focused on rooting out a particular issue, such as bribery in surgery waiting lists, or adopt a multi-pronged approach to focus on the entire service.”

On multistakeholder engagement:

“Mobilizing citizens and stakeholders and strengthening their hand through greater project transparency and openness can build momentum, and change the political economy and cultural considerations that have allowed corrupt practices to happen.

On ‘Govtech’ (which includes the mainstreaming within the public sector of digitization, big data,  cloud computing, artificial intelligence, machine learning, biometrics and digital payments):

“The broadening and deepening of global digitization of governments and citizens is changing the face of public sector governance and its impact on anti-corruption.”

If the ‘govtech’ approach is too-narrowly tech-oriented, the result could be little more than:

 “mere window-dressing… only a veneer of government action…..The traction of digital technologies in reducing fraud and corruption depends on institutional context. Any system will only be as good as the practices that complement it….. Impactful reforms usually require a combination of several layered or sequenced interventions.”

“Framing technology-supported reforms as a public services delivery agenda, rather than an anti-corruption crusade, may  be a more disarming approach in light of the existence of the vested interests benefiting from corruption.”

Two especially evocative cases illustrate how a sector-specific focus, multistakeholder engagement and govtech  can reinforce one another:

Empresas Publicas de Medellin, a municipally-owned infrastructure utility in Colombia:

“EPM proactively makes itself accountable to citizens and shareholders through a series of transparency initiatives, digitization, and customer-engagement practices. Citizens in the community consider themselves to be the real owners of the company and look out for, and protect it, from interference that could be detrimental to its purpose, including a surveillance committee to observe and control the mayor’s decisions and actions that could negatively affect the company.”

“In 2018, in an annual Citizen Perception Survey,  88% of the Medellin population had a favorable image of EPM…. In 2019, EPM created a new communication channel called ‘Transparent Contact’ where citizens could report acts of fraud and corruption involving officials and contractors through a web page, a free hotline, email or fax.”

The National Health System of Ukraine, established in early 2018:

“The NHSU includes an eHealth system for digital health records and reimbursement, initially developed and tested by Transparency International Ukraine… ‘Money follows the patient’ – all public facilities and any private facilities requiring or desiring public financial support had to sign up. Patients signed a declaration with their practitioner, presenting official documentation. Declarations were confirmed using patients mobile phones connected to their registered addresses. Signed declarations underwent central vetting using algorithms to prevent fraudulent or multiple submissions.

“An Affordable Medicines Program provided patients with information about which medicines were covered by the NHSU, and which were not. This increased overall access to medicines and led to a huge reduction in patient co-payments. In July 2019, over four million electronic prescriptions were filled under the NHSU reimbursement program, and over 3 million e-prescriptions issued by pharmacies were reimbursed.”

An October 2018 poll found that 7% of polled patients paid a bribe, compared with 20% in August 2017.”

A focus on the practical is a welcome corrective to hubris.  But, in a world hungry for hope, is it enough? The report’s answer is clear:

“The evidence is overwhelming  that sector specific interventions alone are generally insufficient. Corruption at the sectoral level may flourish because of broader social norms and a governance ecosystem that either tolerates or encourages corruption.”

Though it is not the central focus, there are some tantalizing hints in the report as to what  might  be a mode of engagement capable of making the whole greater than the sum of its parts:

 “Through a sustained and broad-based movement, country examples demonstrate that change can happen at both the project and society level…. What is needed above all, is the commitment of all parties to engage proactively in the fight against corruption through collaboration, innovation and mutual trust”

More from me some other time on a deliberative, coalition-building, norm-strengthening approach to governance work.


How did the World Bank’s October 2020 anti-corruption report manage to avoid the chronic disease of descent into cliché that all-too-often afflicts reports of this kind?  

In my experience, the best of the World Bank is to be found in engaging individually with its staff –  most of those who work on governance bring deep reservoirs of experience, thoughtfulness, professionalism and commitment to their efforts. In recent years, governance practitioners have been sobered, humbled.  Paradoxically, this may have opened up space. Perhaps the loss of center stage has made possible a shift in the balance between wheat and chaff.  As the great singer-songwriter Leonard Cohen put it in his song, Anthem:  

Ring the bells that still can ring.
Forget your perfect offering.
There is a crack, a crack in everything.
That’s how the light gets in.

My congratulations and appreciation to all involved in this fresh, hope-evoking publication.