Overview and approach
1.1. This submission makes detailed comments on two areas of interest to the select committee, where we feel we have sufficient evidence to provide recommendations on DFID’s programming. Our submission is based on the evidence and cross programme learning gathered over years of implementation of sectoral projects in governance, education, private sector and extractive industries and mainstream anti-corruption projects.
1.2. Corruption acts as a major barrier to poverty reduction by undermining service delivery, depressing revenues and discouraging private sector investment (both domestic and foreign). We strongly believe that addressing corruption overseas is beneficial for British interests, as it creates a more conducive environment for British Foreign Direct Investment, as well as safeguarding UK aid spending. Thus, HMG’s increased focus on anti-corruption work has the potential to improve development outcomes that will both contribute to DFID’s overall poverty reduction aims, as well as help to advance British commercial interests abroad.
1.3. A zero tolerance policy towards corruption is impractical and unrealistic, as it ultimately precludes the implementation of projects in low income countries, most of which are transitioning from a patronage basis to full rule of law. We thus recommend a more nuanced and pragmatic approach, one that is informed by political economy analysis and fully cognisant of the underlying drivers of corruption. We believe that anti-corruption work achieves best results in sectors or institutions where there is a significant appetite for reform and where the anti-corruption agenda is aligned with local interests and politics. Furthermore, a realistic, targeted and incremental approach to anti-corruption represents a good Value for Money (VfM) strategy because it leads to what Daniel Kaufmann terms the ‘300% long-run development dividend of good governance’, suggesting that the opportunity cost of failing to tackle corruption can be quantified and is extremely high.
1.4. We realise that working in endemically corrupt environments poses risks to both DFID and service providers. However, accepting corruption as a feature underpinning local power structures and understanding the underlying factors driving it, puts service providers in a better position to respond to it in a pragmatic, effective, and risk-conscious fashion. Further, anti-corruption programmes can provide valuable information, analysis and know-how to inform the design and delivery of DFID’s wider portfolio, thus enhancing development outcomes and safeguarding DFID’s funds.
1.5 When DFID funds are channelled through multilateral institutions which place the funds in the hands of developing country bodies – who then perform the contracting and contract administration – the risk of corrupt practices being encountered is much higher. Few instances will come to light because it is not in the interests of the affected party to complain. That would simply result in them not being paid at all.
Question 1: Should DFID have a zero tolerance policy towards corruption in the countries where it is working or is a more nuanced approach needed to tackle corruption over the long-term? How can DFID manage the risks associated with corruption and reconcile them with its value-for-money agenda?
2.1. Needless to say, a zero tolerance policy approach to corruption should always be applied to service providers. Yet the attainment of development goals and value for money in DFID target countries calls for a nuanced approach. Corruption is a feature of countries transitioning from patronage systems to full rule of law, as well as a characteristic of the governance systems in many fragile and conflict-affected (FCAS) jurisdictions. In our view, DFID’s objective should be to help these countries take an anti-corruption trajectory, rather than adopt an absolutist ‘zero tolerance’ stance. The existence of corruption is better treated with pragmatism and it should be taken into account for the design and delivery of all development projects.
2.2. Appreciating the political economy and role corruption fulfils in developing countries, as well as understanding the political incentives for reform/anti-corruption work is critical. The focus should be on areas where an alignment between politics and the anti-corruption agenda exists; whatever progress is achieved in these areas can then be leveraged to strengthen anti-corruption institutions for when the political economy is more conducive to it. Also, the evidence suggests a targeted and incremental approach can yield results even in contexts where corruption is endemic. These in turn can gradually translate into improvements leading to better service delivery and poverty reduction, which should be the ultimate objective of any anti-corruption intervention.
2.3. As the case of Uganda, where corruption is the norm, demonstrates, progress in certain sectors and institutions is possible, namely the Ministry of Finance, Planning and Economic Development (MoFPED) and the Uganda Revenue Authority (URA). Public Finance Reforms led by MoFPED and supported by DFID resulted in the elimination of more than 9,000 ghost workers, which saved the GoU more than $22 m USD. Similarly, the donor supported institutional transformation of URA led to a 317% increase in revenue collection over the last decade. The establishment of a robust internal compliance unit also contributed to significantly reducing the incidence of bribery in tax administration. According to Transparency International’s East African Bribery Index, corruption incidence in tax services decreased significantly between 2013 and 2014. This is reflected in the scoring improvement from 20.1 in 2013 to 14.5 in 2014 (this is on a scale from 0 to 100, with a score of 100 being the poorest).
2.4. The sections below will provide further evidence to support a nuanced, targeted and incremental approach to anti-corruption using ASI’s programme experience in the extractives sector (Mongolia, Burma and Afghanistan), education (Kenya and Pakistan) and private sector development and infrastructure (Nepal).
Extractives sector case study – Mongolia, Burma and Afghanistan
2.5. It is well established that natural resources like oil and mining tend to distort politics of wealth allocation, and breed corruption. In extractive governance work, ASI has successfully taken a context-sensitive approach underpinned by political economy analysis to establish the entry points for effective reform, transparency and anti-corruption work. This led to, for example, the publication of over 300 mining and petroleum contracts in 2013 by the Ministry of Mines and Petroleum in Afghanistan. To DFID’s credit, the value of political economy analysis in oil and mining has been increasingly understood and adopted.
2.6. ASI’s work supporting extractives revenue transparency in Mongolia, Burma, Afghanistan provides a good example of how transparency and anti-corruption gains in the extractives sector have leveraged cross-sector benefits and been considered for replication in other sectors of the economy. In October 2014 Mongolia, with funding from the EBRD and technical assistance from Adam Smith International, launched an eReporting portal for extractives revenue. Through this portal, figures for environmental payments are now also made available for public scrutiny for the first time. Mongolia has just received an award from the EITI Chair, Claire Short, for implementing the system. In Myanmar – with analytical support provided by ASI and funded by the World Bank, the Government is considering adapting Extractives Industry revenue reporting/transparency standards to the national hydropower sector.
2.7. If DFID is to take a more nuanced and realistic approach to corruption, its real challenge would be to reconcile it with the political and reputational risk this stance bears at home. However, shying away from dealing with corrupt sectors because of such worries would not be a good course. For example, inaction in seeking to improve the governance of the extractive industries sector due to concerns about corruption will exacerbate social, environmental and economic risks.
Education case studies – Kenya and Pakistan
2.8. The education sector is particularly prone to corrupt activities in many DFID target countries. It is characterised by high government spending, with resources being often distributed through multiple administrative levels, and by poor monitoring mechanisms from the central government. Even in decentralised settings it is usually the central government that controls key areas such as budget oversight, recruitment of teachers or payroll. Corruption in education is particularly detrimental to development outcomes. First, it has long-term negative effects on a country’s social, economic and political trajectory; second, it affects the poor disproportionally, who without the access to education have little chance to escape poverty traps.
2.9. In 2009, a fiduciary risk assessment by the Kenyan Ministry of Finance, and an extensive forensic audit of the Kenya Education Sector Strategic Plan (KESSP), discovered substantial financial mismanagement and fraud in the Ministry of Education. Donor funding was suspended and reform initiatives were paralysed. Consequently, DFID’s support has been centred on providing support through other means than general or sectoral budget support. The Kenya Essential Education Programme (KEEP) was thus designed to bypass central government structures. It would not represent a good investment or value for money to channel funds towards the more traditional supply side of education until concrete steps are in place to prevent future leakages. Further, the chaotic state of education management at the central, county, district and school level mean that, even if the system was transparent, inefficiencies in the system would prevent support reaching beneficiaries.
2.10. However, not channelling funding through the government budget does not mean not engaging with the government system. It makes sense to engage with the state system as much as possible to align government and programme goals, while looking towards a time in the not too distant future when the state education system is a transparent, efficient and effective way of delivering education in Kenya. In particular, the development of National Education Sector Plan represents an opportunity to provide advice and guidance in the early stages of a five-year plan, before priorities, work streams and budgets are fixed. KEEP also engages with the local administration level government structures through matching funds payments by County Governments.
2.11 In Pakistan teacher appointments in state schools are sometimes made on the basis of patronage, which contributes to poor educational outcomes as those teachers are often absent. In the Khyber Pakhtunkhwa province in North-West Pakistan the DFID KESP project, which we implement, has supported the provincial government in addressing this by outsourcing teacher recruitment to the private sector, enabling the recruitment system to become merit-based and free from political interference. Further, the programme helped the provincial government implement an ambitious monitoring programme, the results of which are publicised on the internet. Over 28,000 schools are monitored every month, which has resulted in £124k in salaries recovered from absentee teachers and an extra 6,000 teachers being present in Khyber Pakhtunkhwa schools in the first year alone. This is a good example of how transparency – shining a light on the problem – can help prevent corrupt practices.
Private Sector case study - Nepal
2.12. As noted by the IDC in its recent report, corruption is endemic there, which presents a major challenge for DFID. Recently, Nepal slipped to 126 out of 175 countries in Transparency International’s global corruption perceptions index. Bharat Bahadur Thapa, President of Transparency International Nepal, remarked that “corruption really is a serious problem, especially in politics, education, health and in the delivery of services by public servants”. His organisation runs an integrity award for individuals who have made a positive contribution to fighting corruption, but this year they could not find anyone to give it to. Whilst not all public sector organisations are corrupt, many of them are. In these circumstances it does not make sense for DFID to provide budget support, either general or sectorial, to the Government of Nepal (GON). This is also because Nepal above all needs transformational change rather than pound for pound additional funding of existing public services, most of which suffer from systemic problems.
2.13. However, there are a range of approaches which have been shown to work effectively to overcome corruption. DFID’s Nepal Market Development Programme is designed and implemented to minimise any risk of corruption or misuse of funds, while simultaneously achieving success and delivering value for money for DFID. Rather than working directly for the government, the programme instead partners with private firms and community organisations. There is a transparent procurement policy and all partnerships are set up with clear release conditions. Partnership Agreements are then subject to ongoing evaluations and to ensure that all spending is accountable. The programme adapts to changing circumstances to ensure that all DFID funds are safeguarded and the value maximised. For example, NMDP was initially set up with five implementing partners managing interventions. However, after two years the capacity of the core-team had been built up to the point where it made more sense to move towards the direct delivery of interventions. This has worked both to allow closer scrutiny of spending and to reduce administration costs.
2.14. With DFID assistance Nepal has recently negotiated the two largest investments in its history in a process that was notable for its transparency and compliance with international standards. The negotiations embedded a process for the other mega hydro-export deals that are scheduled to follow. Previous cases were often scuppered by competing rent-seeking efforts on the part of politicians/bureaucrats. Avoidance of corruption rested on two major factors: (i) designing an international project development agreement (PDA) and gaining government approval for it (the terms of this agreement then had to be applied, making corrupt arrangements much more difficult to be inserted); and (ii) building up a professional institution, the Investment Board, staffed by a properly remunerated team, who were not susceptible to corrupt practices. Thus in a highly corrupt environment one option is to create a new, non-corrupt institution that sits outside the general system. However this is only worth doing when the rewards are sufficiently great to justify the costs. Moreover, preventing that institution over a period of many years from becoming subject to corrupt forces can be a challenge.
2.15. Leading by example is another approach. As part of the Centre for Inclusive Growth (CIG) programme, DFID financed a solar street light scheme in Pokhara, Nepal’s principal tourist city. This municipal level public-private partnership scheme sets standards for honesty and transparency, and is now being rolled out by the Alternative Energy Promotion Centre (AEPC) in other municipalities across Nepal. Project implementation was delayed by a year because of the necessity to resist efforts to subvert the process to facilitate rent-seeking. The Pokhara example is well known to the AEPC, other municipalities and local business associations, helping to demonstrate the possibility of clean procurement processes.
Question 3: What should the balance be between seeking to tackle corruption top down at institutional level and bottom up at the grass roots? What works and what is not working as well and why?
3.1. ASI has experience implementing accountability programmes (including anti-corruption) with associated demand-side interventions (CSOs, social accountability interventions, citizen engagement, etc.). In general, there is insufficient evidence of the direct impact of demand-side approaches in terms of reducing corruption, and that demand-side solutions are more effective than the exclusively supply-side, technocratic ones. Though we believe that marrying supply and demand-side is considered to be best practice, we need to be realistic about the incentives which invite particular solutions. In many development contexts attempts to engage the civil society where the right mix of incentives is not in place may yield poor results. Thus, we believe that interventions need to be cognisant of the political economy drivers of accountability and anchored in sectors where there is an alignment of political incentives and the anti-corruption agenda. For instance, in political economy environments characterised by clientelism and rent-seeking, collective or public interest action by civil society face constraints. The in-depth diagnostic of the relationship between the citizens and politics must serve as a basis to determine the viability and efficacy of demand-side interventions in a given context.
3.2. While we don’t yet have conclusive evidence of demand-side interventions being effective in Uganda, general conclusions can be drawn. The evidence seems to suggest that the most productive/constructive relationship between CSOs and the GoU is currently the one between the the Ministry of Finance, Planning and Economic Development, (MoFPED), where the incentives for accountability exist, and CSO budget advocacy group. While key Government ministries, agencies, and departments (MDAs) like MoFPED, the Public Procurement and Disposal of Public of Assets Authority (PPDA), and Inspectorate of Government (IG) have productive relationship with CSOs working on budget transparency and procurement issues, so far they have shown a reluctance to work with CSOs focusing on oil governance issues.
3.3. As demonstrated by President Museveni’s strong negative reaction to the debate over the contentious bill to manage Uganda’s oil industry, some CSOs are increasingly perceived by his administration as a threat. CSOs opposing a clause that creates opportunities for mismanagement and corruption – it grants the oil minister the sole authority to grant and revoke licenses and negotiate agreements – were accused of defending spurious and foreign interests.
3.4. An important aspect to bear in mind is that non-state actors also include the private sector, not just civil society, especially in sectors with obvious commercial gains. ASI has played a prominent role in the implementation of the Extractive Industries Transparency Initiative (EITI) across Asia Pacific, South-East Asia, Central Asia, and Africa for over six years now. The EITI is a voluntary initiative that combines external pressure for more transparency in extractives (by virtue of it being a global initiative) with internal pressure for change (by establishing national Multi-Stakeholder Groups of civil society organisations, the private sector and government).
3.5. The evidence suggests that EITI model can achieve real results in terms of reducing corruption, because it requires countries to create multi-stakeholder institutions which enhance accountability. In our view, the EITI model has clearly demonstrated the value of a multi-stakeholder, inclusive approach to transparency and anti-corruption in development work.
3.6. However, anti-corruption gains at scale have not been achieved automatically – our experience shows that there is a strong case for supporting stakeholders (e.g. CSOs and governments) continually in order to realise the benefits from such initiatives, and to scale up the impacts they have.
3.7. Thus, the direct causal relationship between demand-side approaches and their impact on reducing corruption cannot be automatically assumed. The implementation of bottom up initiatives faces many governance and political economy constraints, which may limit their impact. The fundamental issue to address when considering any demand side intervention is about the underlying conditions that are likely to bolster their effectiveness, most crucially the extent which an alignment of incentives between politics and the anti-corruption agenda exists.
1