Skip to main content Skip to navigation

Governmental Reform in Developing Countries: External Conditionality versus Peer Pressure. The Case of Kenya

by Carole Biau and Julie Biau, Department of Economics, University of Warwick [1]



This paper investigates whether World Bank and International Monetary Fund (IMF) conditionality is an appropriate tool for improving governance in developing countries, and what balance of government ownership and donor conditions is needed to achieve meaningful politico-institutional reform. Taking Kenya as a case study, we contrast the effectiveness of current alternatives to the International Financial Institutions’ (IFIs) past Structural Adjustment Programmes. These include a conditionality-based initiative orchestrated by the IMF and World Bank – the Poverty Reduction Strategy Paper and Economic Recovery Strategy for the creation of employment and welfare – and a regional, conditionality-free process, the African Peer Review Mechanism (APRM). We thus attempt to assess whether Kenya’s government can move beyond conditionality and combat poor governance on its own initiative.

Our results suggest that governance-related conditionality is not optimal for promoting institutional reform. IFIs must re-conceptualise governance as a broader socio-political phenomenon, of which financial management and corruption are only one aspect. Reform must also be better internalised by recipient countries. Meanwhile, whilst the APRM solicits greater input from Kenyan stakeholders and better addresses country specificities, it cannot realistically function without binding mechanisms. We also find that government commitment is central to successful institutional reform. We advocate stimulating political will through a peer-consultation initiative inspired by the APRM, but where civil society, external institutions and actors continent-wide participate in defining binding conditions.


Following evidence of fraud in Kenya’s December 2007 presidential elections, this paper’s conclusions (written in September 2007) have been revised in an Afterword (written in January 2008).

KEYWORDS: Governance, Kenya, Corruption, World Bank, IMF, APRM, Conditionality



When the United Nations first used the term ‘governance’ some twenty years ago, many were reluctant to accept this coinage into the political jargon. Today, however, this has become the buzzword of the international community. Good governance is seen as ‘critical to the development process and to the effectiveness of development assistance’ (IDA 12, 1998, cited in Kapur and Webb, 2000: 3). If poor governance exacerbates the economic stagnation of many low-income countries, how can it best be improved? Is the conditionality advocated by International Financial Institutions (IFIs) an appropriate tool for promoting such reform? What balance of government ownership and donor conditions is needed to achieve meaningful change in developing countries?

This paper attempts to answer these hotly debated questions within the context of Kenya. Kenya is a particularly suitable country in which to analyse the effectiveness of governmental reform. Firstly, since its independence in 1963 it has been plagued by poor governance and endemic corruption, consistently ranking among the lowest ten percent of Transparency International’s Corruption Perceptions Index (NEPAD Kenya Secretariat, 2007: 166); the stakes of governmental reform are thus particularly high. Secondly, Kenya’s reputation as a ‘hub of stability’ in East Africa has allowed it to concentrate donor funds and reform efforts of which the country is traditionally tolerant. Kenya thus combines a clear necessity for governmental reform with high potential for the success of such initiatives. Whilst the conclusions drawn from the Kenyan case may not be generalised to all developing countries, they can shed light on the reform paths of other ‘high-potential’ African states in which clientelism poses a similar burden.

Our work develops the findings of five weeks’ research carried out in Nairobi in April 2007. We contrast two international projects geared towards governmental reform: a conditionality-based initiative under the auspices of the International Monetary Fund and World Bank, the Poverty Reduction Strategy Paper (PRSP, reformulated as Kenya’s Economic Recovery Strategy, ERS); and a regional, conditionality-free process, the African Peer Review Mechanism (APRM, conducted under NEPAD, the New Partnership on African Development). Our research relies on interviews with actors directly involved and affected by these reforms, a list of whom can be found in appendix 2. Perception data gathered through interviews are a valid measure of a country’s institutional quality: in the case of corruption, it is the public’s perception of corruption – not the actual corruption level – that informs behaviour. As Kaufmann et al. (2007) explain, ‘when citizens view the courts and police as corrupt, they will not want to use their services, regardless of what the ‘objective’ reality is’ (Kaufmann et al., 2007: 319). Gathering relevant stakeholders’ views is a widespread approach to measuring progress on governance (Kaufmann et al., 2007: 318).

Yet one must note that interviewees often produced contradictory answers, and may not have been utterly sincere due to their positions in the Kenyan administration. We are aware that a quantitative investigation of ERS and APRM achievements would be a useful complement to our evaluation. While an econometric analysis of the determinants of good governance is beyond the scope of the present study, we can suggest several data sources for further research. Transparency International’s Global Corruption Barometer (GCB) and Corruption Perceptions Index (CPI) are composite indices of corruption perceptions indicators and assessments by business people, country analysts and service organisations (TI, 2007: 314, 324). The Kenya Anti-Corruption Commission provides statistics on the value of recovered properties and prosecution rates for corruption cases (KACC, 2005-2006: 17). Perhaps most significantly, the World Bank is currently developing indicators of ‘a nation’s commitment to monitoring enforcement and realisation of anticorruption goals’, which should facilitate future evaluations of governance reforms (TI, 2007: 310).

Being based mostly on qualitative data, this paper therefore does not intend to produce empirical, widely applicable policy prescriptions; rather, we hope to highlight new areas of study by innovatively contrasting the PRSP and the APRM. We also aim to determine which balance of donor influence and government ownership is optimal for Kenya’s reform. Our findings suggest that both programmes have been hampered by poor political will in the Kenyan government; merging these approaches may more effectively encourage sustainable, country-led reform.



Governance refers to the effective rule of law, accountability, public participation, and transparency in the management of the public realm (World Bank, 1998, cited in Mutizwa-Mangiza, 2006: 10). The problem of poor governance is especially pressing in less developed countries, where institutions in which political competition and public scrutiny of the executive can occur are weak. Yet sound institutional frameworks are crucial to promoting economic growth via investment, entrepreneurship, and innovation. Unless such frameworks are built to tackle corruption within governmental bodies, ‘prospects for LDC development will remain very poor indeed’ (Clay, 2007: 4).Kenya’s notoriously poor governance is clearly reflected in its corruption record, which continues to make international headlines: of 18 seriously flawed contracts exposed in 2006, 13 dated from 2002-2003, the period which yielded power to the present government (Clay, 2007: 3). The cost of grand corruption to the Kenyan Treasury in 2006 roughly equalled development aid inflows for that year, leading to the common belief that ‘the best business in Kenya is corrupt business’ (Clay, 2006: 3). Indeed, Transparency International’s Kenya Bribery Index 2007 finds little improvement in the public’s corruption perceptions over recent years (Figure 1; TI, 2007: 8).


Fig 1: Kenyan public perceptions of corruption

Data source: TI Kenya Bribery Index 2007
[Full-size version]


Corruption involves the misuse of public funds by government officials for their personal benefit, and voluntary or forced bribery by citizens to obtain services which they should be entitled to. In Kenya, corruption in public institutions has been a telling indicator of wider governance-related problems, such as lack of democratic space for Kenyan citizens and insufficient government interest in political reform. We therefore use measures of governmental corruption as a proxy for the quality of Kenya’s governance.

From 1978 until 2002, Kenya was led by President Daniel Arap Moi of the Kenya African National Union (KANU). While the state had a healthy economy and a relatively strong institutional framework at independence, corrupt practices became commonplace (Mutizwa-Mangiza, 2006: 15). Poor fiscal management limited effective government investment, and public distrust of the executive hindered private investment. Principally due to ‘weak governance, corruption, and inadequately coordinated […] government actions’, the economy’s 2001 growth rate was the lowest of the post-independence era, at -0.3% (Kiringai, 2001: 1).

The election of President Mwai Kibaki of the National Rainbow Coalition (NARC) in 2002 revived hopes of political transformation. But the NARC government has also disappointed expectations in its battle against corruption. TI’s 2006 CPI still rates Kenya at 2.2, up from an average of 2 in pre-election years (TI, 2006: 330). The CPI being on a scale of 0 to 10, where 0 is highly corrupt, this reflects very little improvement. Kenyans still perceive the government office as the country’s most corrupt body (TI interview). For former British High Commissioner to Kenya Sir Edward Clay, ‘the government which promised […] a war on corruption and a new constitution, has instead embraced corruption and denied its citizens constitutional reform’ (Clay, 2006: 4). The government may be deploying insufficient efforts towards improving governance.



Conditionality refers to the setting of conditions by donor agencies and international institutions; recipient countries must meet these conditions for successive tranches of funding to be disbursed. Conditionality was the basis of the World Bank’s and IMF’s Structural Adjustment Programmes of the 1980s; it is widely recognised, however, that the SAPs were not the panacea their proponents expected them to be. The failure of traditional conditionality is partly attributed to deficiencies within the IFI programmes themselves: Joseph Stiglitz suggests that IFI staff often use ethnocentric, ‘bad economics’ in their recommendations (Stiglitz, 2007: 13), for Ngaire Woods the hierarchical rigidity of IFI structures impedes programme innovation (Woods, 2006: 104). Yet recipient country characteristics, such as the presence of good governance institutions and rule of law, also impact the adjustment process. Indeed, on the basis of 220 World Bank adjustment programmes, Dollar and Svensson (2000: 905) find ‘no evidence that any of the variables under the World Bank’s control affect the probability of success or failure of an adjustment loan’, which depends on domestic political-economy forces. This conclusion shakes the very foundations of traditional IFI policies, and suggests that the omission of governance-related clauses in the SAPs may have heavily contributed to their malfunction.

Conditionality is further criticised for undermining state sovereignty by rendering governments accountable to external bodies rather than to their own people. ‘A nation’s desperate need for short-term financial help’, Buira argues, ‘does not give the IMF the moral right to substitute its technical judgements for the outcome of the nation’s political process’ (Buira, 1998: 57). Such arguments depict developing countries as the powerless recipients of external policymaking – they must however be moderated in the case of Kenya, which is not aid-dependent: donor support represents less than 2% of the national budget and has not been factored into budget planning for the last two years (Interview with European Commission). Nonetheless, as interviewed NEPAD representatives noted, the ‘immoral and condescending’ nature of conditionality provides little motivation for change within recipient countries.

The failure of traditional IFI adjustment programmes based on conditions of a solely macroeconomic nature, coupled with findings highlighting the importance of good governance for successful reform, has dramatically shifted the focus of conditionality. Since the 1990s, the IFIs have undergone what Woods calls a ‘second-generation reform’ (Woods, 2006: 129): ‘good governance has become enshrined in the[ir] commandments’ (Kapur and Webb, 2000: 3). Yet an evaluation of these institutions’ current programmes in Kenya raises questions as to whether conditionality is appropriate for tackling questions of political reform.

1. Introducing governance into conditionality: poverty reduction strategy papers

In September 1997, the World Bank adopted a policy statement that ‘corruption should be explicitly taken into account in country risk analysis [and] lending decisions’ (World Bank, cited in Kapur and Webb, 2000: 3). IFI conditions have since been repackaged as Poverty Reduction Strategy Papers (PRSPs), with a novel focus on governance, transparency, accountability and institutional reform. In Kenya, the PRSP’s governance-related clauses include downsizing of the public sector, reduction of the government’s bloated wage bill, setting up the Kenya Anti-Corruption Commission (KACC), and passing the Economic Crimes Bill and Public Officers Ethics Act.

The PRSP’s most fundamental innovation lies in its focus on government ownership of reforms. At their September 1999 annual meetings, the World Bank and IMF proposed that all their conditional lending be based on ‘country-owned’ poverty reduction strategies (Levinsohn, 2003: 121). As Kapur and Webb (2000: 8) explain, ‘reforms rarely succeed unless a government shares the conviction that they are essential’. PRSPs therefore request that plans of action be drafted by the recipient governments before submission to the Bank and Fund for approval. Only then can countries apply to the IMF’s advisory board for funding (Levinsohn, 2003: 121; Interview with the IMF). Solutions are therefore designed through improved consultation with the recipient country and with reference to its particular circumstances, reflecting a genuine change from traditional lending initiatives.

2. Evaluating the ERS / PRSP effectiveness

In Kenya, although a PRSP was completed in 2000 under the Moi regime, funds were not disbursed for the programme due to strained relations with donors. With the 2002 election of the NARC coalition, however, the PRSP was reformulated as the Economic Recovery Strategy for the creation of employment and welfare (ERS) and funding was renewed. We will investigate whether the ERS really is a government-owned and consultative process (Section 2.1) and whether it has been effective in improving governance (Section 2.2).

2.1 a. Government ownership

Interviews with donor agencies and Kenyan ministerial staff suggest that the ERS is truly country-owned. A European Commission representative had ‘never heard complaints that [the ERS] is donor-driven’. Similarly, a Ministry of Planning employee explained that donors ‘supported what the government was already doing without interference’ and felt that his government fully controlled the action plan. As underlined by an IMF representative, the PRSP conditionality assesses the country’s progress according to its own benchmarks. Thus, the IMF requires evidence that the country is keeping to its intended plan of action and disburses loan tranches accordingly. Ideally, the Fund bases its conditions on entirely government-generated targets (Interview with the IMF). Country-ownership was also a clear priority for World Bank representatives.

Yet despite this insistence on country-generated guidelines, as put by an IMF representative, ‘that’s rarely the way it actually works out.’ In practice, he explained, countries often make these guidelines ‘as soft as possible’, or set unrealistic targets. While the IMF is ‘not interested in making it as tough as possible,’ it requires markers that clearly indicate that project objectives will be met. An World Bank economist whom we interviewed also stressed the need for setting firm conditions. Indeed, overly optimistic targets had to be revised in implementing Kenya’s ERS, and meeting these has proved problematic: whereas six disbursements should have occurred since 2004, by 2007 none had.

A further difficulty is that many countries lack the qualifications to draft sound poverty reduction strategies (Levinsohn, 2003: 124). Even in Kenya, which has access to high-level economists, the Bank and Fund often intervene to provide technical assistance, which they concede reflects their own policy preferences (interview with the IMF). For Levinsohn, the notion of country-driven strategies may expect too much out of recipient countries. In the absence of sufficient analytical capacity on behalf of the country, he writes, ‘one just gets platitudes and . . . a discussion of what the country thinks the Bank and Fund want to hear’ (Levinsohn, 2003: 125). The IMF representative interviewed acknowledged this possibility. Many practicalities prevent the Kenyan government from taking full ownership of its poverty reduction strategy.

2.1 b. A consultative process

For the ERS to be fully country-owned, it is vital to consult not only the country’s government but also its population. Rather than simply depending on the ministry of finance’s approval, ERS elaboration requires the participation of many ministries, ethnic minorities and other stakeholders (Levinsohn, 2003: 121). The Ministry of Planning explained that throughout drafting the ERS the government engaged in Consultative Group meetings with development partners, followed by regular reunions with the KACC, civil society organisations, and members of the private sector. When asked about its involvement as a civil society organisation in the consultation process, the NGO Transparency International likewise acknowledged that its opinion was frequently solicited.

However, staff members were sceptical as to the impact that their input really had on the final ERS (Interview with TI). This is partly because Kenya lacks a systematic forum that can bring members of civil society together to share their findings, and monitoring by NGOs is disorganized due to poor financial support. This has led to a deficit of participation and to an exclusion of the poor from consultation (Shiverenje, 2005: 31). Additional procedures to institutionalise the stakeholder participation must be established before reform can truly be country-owned.

2.2. Effectiveness of implemented programmes

The ERS’s principal governance-related clauses centre on tackling corruption; accordingly, we evaluate the programme’s effectiveness by observing the outcomes of its anti-corruption initiatives. These include the setup of the Kenya Anti-Corruption Commission in 2003 and the passing of the Procurement Bill, Budget Overlook Paper, Performance Contract, Public Officers Ethics Act, and Public Officers’ Performance Act. Whilst interviewed representatives of the Ministry of Finance and of the KACC insisted on these bills’ effectiveness, critics question whether they are genuine, or implemented simply to secure donor funds. As suggested by European Commission representatives, ‘the government’s intention [at times] seems to be to create a lot of new laws, but not to improve anything.’

2.2 a. The Kenya anti-corruption commission

Established under the Anti-Corruption and Economic Crimes Act in 2003, the KACC’s mandate includes the prevention, detection, and investigation of corruption, as well as law enforcement against corruption practices (KACC, 2005-2006: 4). Since its inception the KACC has faced accusations of incompetence. Its structure is rather limited in terms of sanctioning corruption: out of 7,888 reports of economic crime brought to the Commission in 2005-2006, only 15% fell under its mandate (KACC, 2005-2006: 4). Furthermore, as the Commission itself has no powers of prosecution, actionable cases must be submitted to Kenya’s Attorney General (AG) before legal action is undertaken. As the current AG retains his functions from lengthy service in the Moi regime, he is feared to have vested interests in multiple cases; indeed, no large-scale corruption cases have been brought to court since 2003 (Interviews with TI, IMF). Some even describe the KACC as ‘a total façade’ which ‘does not give the AG enough evidence to prosecute cases’ (Interview with FNSF).

The public perception of the KACC has consequently worsened since its establishment. The KACC’s claims that it is ‘100% independent’ from the government no longer seem credible. As put by an IMF representative, ‘the KACC makes it harder for you to steal, but if you succeed, you’re okay’. In reply to such accusations, KACC staff objects that the institution has significantly progressed in the area of prevention, primarily through civic education. Some of these efforts have paid off: the Kenyan public’s willingness to report bribes has increased from 9% to 14% of respondents since 2005, and citizens’ perceptions of improvement in government institutions have risen by 10% during this time (TI, 2007: 8).

As noted by KACC’s Principal Officer for Prevention, 2005-2006 was the Commission’s first year of full operation with a complete staff, and it may simply be too early for grand corruption cases to have come to court. A Ministry of Planning representative likewise suggested that the public ‘may be rushing to see results without considering the due process’. Such arguments however appear insufficient in the face of growing discontent with this institution.

2.2 b. Passed anti-corruption bills

The second ERS-led initiative in terms of combating corruption has been the enactment of acts and bills. The Performance Contract, Public Officers Ethics Act, and Public Officers’ Performance Act aim to increase incentives for public officers to refuse bribes and to instil a culture of transparency in government offices. Under the Performance Contract, public officers must meet work quotas within a given time period; a Ministry of Planning representative claimed that this Contract has increased administrative efficacy, and argued that the salary increment implemented under the Public Officers Ethics Act has ‘made people want to work more’. He concluded that most colleagues had ‘grown to dislike bribes’ and were now unwilling to take them. European Commission representatives also expressed optimism, and the TI Kenya Bribery Index 2007 notes that, ‘the corporate governance and procurement reforms undertaken […] are paying off’ (TI, 2007: 8).

Yet the effectiveness of the ERS’s anti-corruption clauses remains ambiguous. The Kenya Bribery Index 2007’s survey respondents encountered bribery in 54% of their interactions with public and private institutions, up from 47% in 2005; Kenya’s total bribe burden has increased by 50% over the last two years (TI, 2007: 20). Government interference in the enactment of anti-corruption bills has also raised criticism. The results of the 2003 Public Officials Act, whereby high-ranking government officials had to declare their wealth, were for instance never made public (Interview with FNSF). For TI, ‘these laws are designed mostly to improve the record and to warm up to donors’ rather than to initiate genuine change. As Figure 2 (adapted from the GJLOS Baseline Report 2007) suggests, much of the Kenyan public shares this view: only 16% of respondents believe the government is ‘committed’ to reform’, and 24% claim that it is ‘not committed’. Likewise, the Kenyan public has reservations concerning the effectiveness of the KACC and of ERS-promoted legislation (Figure 3): the majority of respondents think that these efforts have reduced corruption ‘only a little’ over the past year.


Fig. 2: perceived level of government commitment to reducing overall corruption

Data source: GJLOS 2006 Baseline Report 2007
[Full-size version]


Fig. 3: perceived effectiveness of ERS-promoted governance reforms

Data source: GJLOS 2006 Baseline Report 2007
[Full-size version]


An intriguing aspect of the GJLOS Report, which collects data by province, is that optimistic respondents are concentrated in the more economically-active provinces (especially Nairobi and Central), whilst the majority of respondents in particularly poor provinces (such as Nyanza and North-Eastern) report ‘little effectiveness’ or ‘don’t know’. These discrepancies indicate that poorer provinces may be slighted by government efforts, and also reveal lower civic awareness in these areas. Such factors may facilitate the government’s elusion of reform, and may deserve further study.

Our findings thus suggest that, while bills and reforms requested by the ERS have improved transparency and awareness of ‘petty’ corruption in Kenya, high-level corruption may remain undisclosed in the absence of genuine political will. This signals wider dysfunctions within Kenyan governance structures, which the ERS is unable to address.



Many critics claim that the ERS’s impact is low simply because conditionality is an inappropriate tool for monitoring governance. IFIs may not have sufficient experience in the area of governance to adequately address governmental reform in their conditions: as an IMF representative admits, IMF decision-makers are ‘just a bunch of PhD economists […] and how far the IMF should go along the road of governance is an ongoing debate’. More condemnatory, TI representatives concluded that conditionality ‘created avenues for corruption’: the moves required by donors can disrupt ongoing government efforts to regulate corruption, and donor-initiated bodies (such as the KACC) are often too expensive for governments to run realistically. Conditionality may thus deepen the problems it attempts to solve by ‘imposing extra costs on countries and making governments shift their priorities’, eclipsing local initiatives (interview with TI).

Imposing conditions on governance may also generate a backlash from the recipient, as ‘the government does not want to be dictated to’ (interview with APRM Programme Officer). Yet can governance be improved internally and without conditions? Analysing Kenya’s prime alternative to conditionality, the African Peer Review Mechanism (APRM), is key to answering this question.



In 2002, member states of the African Union agreed to the AU Declaration on Democracy, Political, Economic and Corporate Governance; the signatories committed themselves to upholding standards of good governance. The following year, the APRM was set up under the auspices of the New Partnership for Africa’s Development (NEPAD). This self-reviewing mechanism focuses on improving socio-economic development, democracy, and political, economic and corporate governance continent-wide. The APRM is voluntarily acceded to by states and stresses national ownership of action plans; unlike other multilateral approaches, it is internal to Africa and entirely conditionality-free. It also adopts a more comprehensive view of governance, a significant improvement over IFI initiatives. The APRM is seen as ‘a friendly learning process’ through which African countries assess each others’ performances, share experiences, and identify weaknesses and successful practices in improving governance (NEPAD Kenya Secretariat, 2007: 29).

Kenya was the third country of the AU to volunteer for assessment in 2003. A 33-member independent APRM National Governing Council, nominated by civil society stakeholders, was established to oversee the process. The final country report concludes on a note of optimism, describing the government’s August 2005 Self-Assessment Report as ‘the most thorough expression of public opinion to have ever occurred in Kenya’, having drawn on eighteen months of household surveys, public forums and plenary sessions held across the country’s eight provinces (Mutizwa-Mangiza, 2006: 22). The report notes that ‘unlike his predecessor, President Kibaki has demonstrated courage and will to end corruption in the public sector’ (NEPAD Kenya Secretariat, 2007: 246).

Interviewed NEPAD and NGO representatives echoed this enthusiasm, viewing the APRM as a substitute for conditionality-based approaches. It capitalises on competition between African states rather than on uneven donor-recipient relationships, and may thus provide an incentive structure better suited to improving governance than does the traditional ‘carrot-and-stick approach’ (interview with FNSF). The APRM is hoped to ‘succeed . . . where lectures, browbeating, incentives and the rest of the donor paraphernalia have failed’ (Clay, 2006: 2). For the APRM Programme Officer to Kenya, this self-initiated project indicates genuine will for reform in African countries, and heralds that ‘conditionality is going to be a thing of the past’.



Views of the APRM as a ‘more credible alternative to conditionality’ (Interview with APRM) are however highly contested. For a European Commission representative, the APRM, like the ERS, is ‘a bit of a shopping list’: poorly outlined priorities and the proliferation of clauses and bills ‘make it easy [for the executive] to talk itself out of the situation’ and to exhibit a façade of reform. The APRM’s very logic is also questioned, many being sceptical of ‘the notion of states such as Nigeria, which have had demonstrably notorious governance records, becoming reviewers’ (Mutizwa-Mangiza, 2006: 31). The APRM moreover risks granting African leaders undeserved certificates of good conduct. As Clay notes, ‘governments will not be too hard on fellow-governments […] who knows when one government will not need a ‘Get Out of Jail Free’ card from the others?’ (Clay, 2006: 5). Unless all participants are genuinely committed to combating corruption, the APRM’s structures may be overly optimistic.

Since its implementation, Kenya’s APRM process has indeed suffered enough controversy to suggest that the necessary political will is severely lacking. The mechanism’s effectiveness was particularly shaken in July 2005, when Ms Grace Akumu, former chairperson of the APRM’s National Governing Council, questioned the use of funds and the level of public involvement in writing the APRM review. Ms. Akumu argued that, ‘the government is not supposed to interfere with the review […] the report will have no integrity’ (Mutizwa-Mangiza, 2006: 27). Three APRM officials, including Ms Akumu, were then decommissioned, and Kenya subsequently ‘rubbished’ the 2006 Country Report (Mutizwa-Mangiza, 2006: 12). Critical factual errors in this report, such as attributing the notorious 2003 Anglo-Leasing scandal to the former KANU regime rather than to the NARC government, raise further doubts over the autonomy and reliability of the review process (Clay, 2006: 5).

As an endogenous, voluntary approach to improving governance in Africa, the APRM has promising fundamentals. However, the apparent lack of government commitment to change has significantly undermined its credibility in Kenya. The initiative’s main weakness is the absence of binding mechanisms and its reliance on governments to denounce the very corruption from which they may themselves benefit. Yet given the significant problems with the donor-imposed approaches mentioned above, such a valuable opportunity to move beyond conditionality should not be too rapidly dismissed. As TI representatives remind us, ‘Kenya has just come out of twenty years of autocratic rule […] the opening up of political space will take time’. Which balance of conditionality and government ownership will be optimal for Kenya in coming years ultimately depends on the emergence of genuine political will in the country.



The KACC Annual Report 2005-2006 proclaims that, ‘for the first time in Kenya’s history, the war on corruption is no longer sporadic, half-hearted and reactionary, but consistent’. This assertion reflects the remarkable optimism conveyed by interviewed stakeholders, who generally agreed that Kenya was making determined efforts to improve governance under the NARC coalition. According to a KACC officer, legislations such as the Performance Contract and the Rapid Reform Initiative could not have been introduced in the absence of political will. As expressed by the Ministry of Planning, ‘the will [for reform] is there and it is from the top.’

Accordingly, since the Paris Declaration on Aid Effectiveness (2005), the donor community is gradually reducing the conditions on its funding. The European Commission has begun budget support to Kenya, injecting money directly into government accounts rather than tying aid to specific projects; yet it is so far the only donor to have taken this step. Others such as DFID remain more prudent, believing that political will is not yet strong enough for the Kenyan government to implement change in the absence of direct penalties. Disappointing experiences with the APRM process, the ambivalent conclusions of TI’s latest Bribery Index Report, and the fact that cases of corruption are still being exposed, all reflect this need for caution. Had the Kenyan government truly been committed to reform, the APRM and ERS would have more effectively improved governance.

This study highlights the importance of political will as the prerequisite for the success of any development-geared effort. Many development agencies (especially the UNDP) are today attempting to side-step the political will problem by decentralising their aid and granting local initiatives preference over state-centred frameworks. Studies show that these grass-roots approaches are however of little avail in the absence of political will: elite capture is frequently replicated locally by influential civil society representatives (Platteau, 2003: 4). Lack of commitment is thus not simply limited to the spheres of governmental elite, but is prevalent society-wide. Further sociological research into this phenomenon’s underlying causes would bring valuable contributions to development efforts. Corruption in Africa and parts of Asia is often attributed to a culture of particularistic tribal or family-based (rather than national) solidarity networks, which may legitimate nepotistic redistributions of wealth in the eyes of the public (Courade, 2006: 205). Civic awareness may therefore be the primary tool through which to discourage clientelism and to stimulate local and national commitment to development programmes. Globalisation is also upsetting traditional pluralistic networks, strengthening certain interviewees’ arguments that political will could improve with Kenya’s next generation.



The APRM country report explains that ‘Kenya has many good laws, commissions, programmes and institutions that could make it the best-run democracy in Africa’ (NEPAD Secretariat, 2007: 242). For almost half a century, however, corrupt practices and poor government commitment have prevented this transformation from taking place. Our investigation of governance-related conditionality in the form of the Economic Recovery Strategy suggests that this approach is not optimal for promoting institutional reform in the country. As expressed by Joseph Stiglitz, ‘good policies cannot be bought, at least in a sustainable way’ (Stiglitz, 1999, cited in Kapur and Webb, 2000: 7). IFIs must re-conceptualise governance as a more social, economic and political phenomenon, of which corruption and transparency are only one aspect. These principles must furthermore be internalised by recipient countries: although government ownership is officially built into recent programmes, there must be genuine dedication of the executive to the reform process.

Contrasting conditionality-based lending to the African Peer Review Mechanism highlights the potential of regional alternatives. The APRM solicits greater input from Kenyan stakeholders and is more finely tuned to country-specific circumstances. However, its disappointing outcomes suggest that this programme cannot realistically function in the absence of binding mechanisms. The APRM must be more objective and action-oriented, and must mobilise more independent expertise; linking with the United Nations, for instance, may give it the neutrality and credibility it requires.

We conclude that political will is key to successful institutional reform; conditions and participatory frameworks alone cannot render government bodies fully responsible. If political will is lacking, any reform advice that assumes its presence – no matter how country-specific or inclusive of civil society – may well fail. The ERS and the APRM both clearly suffer from insufficient government commitment. Nonetheless, many interviewees claim that a change in mindset is imminent, and that ‘the new generation has Kenya at heart’ (interview with FNSF). To catalyse this evolution, we advocate using a peer-consultation initiative inspired by the APRM to stimulate political will. This improved assessment mechanism could include civil society and actors continent-wide in the definition of binding conditions, provided that indicators of good governance are also agreed upon with external institutions. Once this stricter, more objective and representative peer assessment is established, the potential for moving beyond conventional conditionality may truly be in place.



Since the writing of this article, Kenya’s NARC government (now PNU, Party of National Unity) has faced its first genuine trial of political will: the holding of presidential and parliamentary elections in December 2007. These elections’ outcome has become sadly notorious: widespread irregularities – to which the Electoral Commission has admitted – granted the outgoing president Mwai Kibaki a fictitious victory over opposition leader Raila Odinga (Orange Democratic Movement, ODM). The stark contrast between the president’s re-election and the ‘vote of no confidence’ placed for PNU in the parliamentary elections (in which the party obtained only 46 of the 210 seats) is clear evidence of such rigging. This complete lack of political integrity was reconfirmed on January 8, with Mwai Kibaki unilaterally appointing seventeen members of his cabinet just hours before power-sharing negotiations were scheduled to be held with ODM (The Monitor, 29 February 2008).

Many have taken the recent constitutional reform instating Raila Odinga as Prime Minister with executive powers to be a positive sign for democracy. However, deep controversy remains over the nomination of Cabinet officials and the sustainability of the PNU-ODM ‘power-sharing’ arrangements. The government has calamitously failed its most important test, and demonstrated that the political will of past years may indeed have been a mere ‘façade’. Sadly, last September’s conclusions were overly optimistic – in its present situation and whilst latter-generation elites remain in power, Kenya is in no measure ready for the ‘move beyond conditionality’.




The primary research for this article was funded by the University of Warwick’s Undergraduate Research Scholarship Scheme.



Fig. 1: Kenyan public perceptions of corruption (TI Bribery Index 2007)

Fig. 2: Perceived level of government commitment to reducing corruption (GJLOS 2006 baseline report 2007)

Fig. 3: Perceived effectiveness of ERS-promoted governance reforms (GJLOS 2006 baseline report 2007)



APRM African Peer Review Mechanism
AU African Union
CAS Country Assistance Strategy
CPI Corruption Perception Index
DFID Department for International Development (UK)
EC European Commission
ERS Economic Recovery Strategy for the creation of employment and welfare
ESAF Enhanced Structural Adjustment Facility
FNSF Friedrich Naumann Stiftung Foundation
IFIs International Financial Institutions
IMF International Monetary Fund
KACC Kenya Anti-Corruption Commission
KANU Kenya African National Union
NARC National Rainbow Coalition
NEPAD New Partnership for Africa’s Development
NGC National Governing Council
PRSP Poverty Reduction Strategy Paper
SAP Structural Adjustment Programme
TI Transparency International
UNDP United Nations Development Programme
UNHDR United Nations Human Development Report


Appendix 1: Ten years of governance initiatives in Kenya

1997: IMF stops ESAF disbursements due to poor progress in governance.
2000: PRSP document completed under KANU regime; lending restarts.
Dec. 2000: Lending suspended due to new evidence of grand corruption.
Dec. 2002: NARC government elected, IMF and WB willing to resume lending.
2003: Setup of Kenya Anti-Corruption Commission (KACC).
Mar. 2003: Kenya joins APRM.
May 2004: ERS (reformulation of the PRSP under the new government) initiated.
Oct. 2004: APRM National Governing Council appointed.
2004-2007: CAS initiated (WB programme based on ERS).
Aug. 2005: Draft APRM programme of action & Internal Self-Assessment completed.
May 2006: Review of APRM Kenya Report at AU summit in Banjul.
Aug. 2007: TI Kenya Bribery Index reveals uneven progress in tackling corruption.
Dec. 2007: Evidence of fraud in Kenya presidential and parliamentary elections, leading to a revision of the conclusions of the current paper.


Appendix 2: List of interviews carried out as part of this research

World Bank: April 14th 7:00 PM, Thigiri Gardens, New Muthaiga, Nairobi.

IMF Kenya: April 17th 12:00 PM, Kenyari Towers 10th floor, Upper Hill, Nairobi.

European Commission: April 16th 11.00 AM, Union Insurance House, Regati Road, Upper Hill, Nairobi.

Friedrich Naumann Stiftung Foundation: April 14th 11:00 AM, United Nations complex, Gigiri, Nairobi.

Transparency International Kenya: April 17th 10:00 AM, Community Centre, ACK Garden House, Nairobi.

APRM Kenya: April 16th 12:00 PM, Liason House, Statehouse Avenue, Nairobi.

Ministry of Planning: April 16th 2:00 PM, Treasury Building, Nairobi.

Kenya Anti-Corruption Commission: April 16th 4:30 PM, Integrity Centre, Nairobi.

DFID Kenya: April 17th 11:00 AM, British High Commission, Upper Hill Road, Nairobi.

Former British High Commissioner to Kenya: June 25th, email interview.

Members of the general Kenyan public.



[1] Julie Biau is currently undertaking an Erasmus year abroad in Universidad Carlos III, Madrid. Carole Biau is currently undertaking an Erasmus year abroad in Sciences-Po Paris. Both Julie and Carole will return to Warwick next year to complete their BSc's in Economics, Politics and International Studies.



Buira, Ariel (ed.) (2003), Challenges to the World Bank and IMF: Developing Country Perspectives, G24 Research Program, London: Anthem Press

Clay, Edward (2007), ‘How Bad Governance Impedes Economic Development’ in Warwick International Development Summit Post-Summit Magazine 2006, Warwick International Development Society (WIDS)

Clay, Edward (2006), ‘The African Peer Review Mechanism: Not Yet Fit For Purpose’, personal communication

Courade, George (ed). (2006), L’Afrique des idées reçues, Collection Mappemonde, Belin

Danaher, Kevin and Muhammad Yunus, (eds) (1994), 50 Years is Enough: the case against the World Bank and the International Monetary Fund, South End Press

De Blasio, Guido and Alberto Dalmazzo (2003), ‘Resources and Incentives to Reform’, IMF Staff Papers Vol. 50, No. 2

De Maria, William (2005), 'The new war on African corruption: Just another neo-colonial adventure', in Gilson, C. H. J. (ed.), Critical Management Studies Proceedings 2005: Critique and Inclusivity: Opening the Agenda, Waikato Management School, Waikato, New Zealand

Dollar, David and Jakob Svensson (2000), ‘What explains the success or failure of structural adjustment programmes?’, The Economic Journal, 110, 894-917

Government of Kenya in collaboration with the International Monetary Fund and World Bank (1996), Economic Reforms for 1996-1998: The Policy Framework Paper

Government of Kenya (2007), ‘GJLOS National Integrated Household Baseline Survey – Data Tables’, Governance, Justice, Law and Order Sector (GJLOS) Reform Programme
Government of Kenya, Poverty Reduction Strategy Paper for Kenya, 2001-2004

Ghosal, Sayantan and Kannika Thampanishvong (2007), ‘Optimal Sovereign Debt Write-Downs’, CMDA Working Paper No. 07/10, Centre for Dynamic Macroeconomic Analysis, University of St Andrews

Grant, Ruth W. and Robert O. Keohane (2005), ‘Accountability and Abuses of Power in World Politics’, American Political Science Review, 99 (1), 29-43

Hughes, Alexandra (2002), Lessons Learnt On Civil Society Engagement in PRSP Processes in Bolivia, Kenya And Uganda: a Report Emerging From the Bolivian-East African Sharing and Learning Exchange, Brighton: Institute of Development Studies (IDS)

Kenya Anti-Corruption Commission, KACC Annual Report 2005-2006

Kanbur, Ravi (2000), ‘Aid, Conditionality and Debt in Africa’, in Tarp, Finn (ed.), Foreign Aid and Development: Lessons Learnt and Directions for the Future, London and New York: Routledge, pp. 409-22

Kanbur, Ravi (2002), ‘IFI’s and IPG’s: Operational Implications for the World Bank’, G24 Technical Group meeting, Beirut

Kanbur, Ravi and Matti Tuomala (2001), ‘Incentives, Inequality And The Allocation Of Aid When Conditionality Doesn't Work: An Optimal Nonlinear Taxation Approach’, Working Paper 2001-11, Department of Applied Economics and Management Cornell University, Ithaca, New York

Kapur, Devesh and Richard Webb (2000), ‘Governance-related Conditionalities of the International Financial Institutions’, G-24 Discussion Paper Series, United Nations Conference on Trade and Development, 6 (August 2000)

Kaufmann, Daniel, Aart Kraay and Massimo Mastruzzi (2007), ‘Measuring Corruption: Myths and Realities’, in Transparency International (2007), Global Corruption Report 2007, Cambridge University Press, pp. 318-23.

‘Kenya: The Agreement’, Kampala: The Monitor, 29 February 2008,, accessed March 1 2008

Kiringai, Jane (2001), ‘Debt and the PRSP Conditionality: The Kenya Case, Preliminary Draft’, Kenya Institute of Public Policy Research and Analysis (KIPPRA)

Knack, Stephen, and Aminur Rahman (2007), ‘Donor Fragmentation and Bureaucratic Quality in Aid Recipients’, Journal of Development Economics, 83, 176-97

Levinsohn, James A. (2003), ‘The poverty reduction strategy paper approach: good marketing or good policy?’, in Buira, Ariel (ed.), Challenges to the World Bank and IMF: Developing Country Perspectives, G24 Research Program, London: Anthem Press

Mauro, Paulo (1995), ‘Corruption and Growth’, Quarterly Journal of Economics, 110 (3), 681-712

Mutizwa-Mangiza, Shingai Price (2006), ‘Good Governance and National Development: a Review of the Implementation of the African Peer Review Mechanism in Kenya’, personal communication

NEPAD Kenya Secretariat (2006), ‘African Peer Review Mechanism: Country Review Report of the Republic of Kenya’

NEPAD Kenya Secretariat (2006), ‘African Peer Review Mechanism (APRM): Popular Version Kenya Report’

O’Brien, F. S. and T. C. I. Ryan (1999), Aid and Reform In Africa: Kenya Case Study, Washington, D.C.: World Bank project on Aid and Reform in Africa

Platteau, Jean-Philippe and Gaspart, Frédéric (2003), Extracts from ‘Disciplining Local Leaders in Community-Based Development’, Belgium: Centre for Research on the Economics of Development (CRED)

Rahman Kahn, Azizur (1993), Structural Adjustment and Income Distribution, Geneva: International Labour Office

Republic of Kenya (2003), ‘Kenya Gazette Supplement: Acts, 2003, The Anti-Corruption and Economic Crimes Act, 2003’, Nairobi: Government Printer

Shiverenje, Hudson (2005), ‘What Happened to the PRSP in Kenya? The Role of Politics’, in National Council of NGOs, Facts on the Economic Recovery Strategy: Civil Society Perspective, Participatory Learning and Action 51, pp. 27-29

Sobhan, Rehman (2005), ‘Democratising the MDG Agenda’, in Achieving the Internationally Agreed Development Goals, United Nations Department of Social and Economic Affairs, United Nations Publications, pp. 106-11

Stiglitz, Joseph (2007), Globalisation and its Discontents, London: Penguin Books

Toh, Kiert (2004), ‘Conference on Agricultural Competitiveness for Economic Growth and Poverty: Speech by Dr. Kiert Toh, USAID/Kenya Mission Director’, Nairobi

Transparency International (2007), Global Corruption Report 2007, Cambridge, Cambridge University Press

Transparency International Kenya (2007), ‘The Kenya Bribery Index 2007’

Woods, Ngaire (2006), The Globalizers: the IMF, the World Bank, and their Borrowers, Ithaca: Cornell University Press

United Nations Development Programme (UNDP) (2006), Human Development Report 2006, Ch. 3: ‘Aid for the 21st Century’

UNDP Kenya (2005), Kenya Development Cooperation Report 2004

Van de Walle, Nicholas (1994), ‘Review: Adjustment Alternatives and Alternatives to Adjustment’, African Studies Review, 37 (3), 103-17

World Bank (2004), ‘A More Equitable, Prosperous and Competitive Kenya: Country Assistance Strategy (CAS) 2004-2007’, World Bank Group


To cite this paper please use the following details: Biau, C. and Biau, J. (2008), ‘Governmental Reform in Developing Countries: External Conditionality versus Peer Pressure. The Case of Kenya', Reinvention: a Journal of Undergraduate Research, Volume 1, Issue 1, Date accessed [insert date]. If you cite this article or use it in any teaching or other related activities please let us know by e-mailing us at Reinventionjournal at warwick dot ac dot uk.

© Reinvention: a Journal of Undergraduate Research (2008). Full copyright remains with the author.