The Political Economy of Aid and the Good

The Political Economy of Aid and the Good Governance Agenda in Africa Accra Residential School on Governance and Development CARLOS OYA Development Studies, SOAS, University of London Email: co 2@soas. ac. uk Accra, 8 th May 2013 1

Outline • The aid ‘complex’, ‘aid effectiveness’ debates and some background • Aid and state capacity ‘de-building’ • Conditionality, policy space and ideology • The pitfalls and contradictions in the nexus aid-governance 2

Some key issues in the aid-governance nexus 1. Aid and state formation 2. Aid as factor affecting the nature of state institutions and practices (rentier, neopatrimonial, developmental, etc. ) 3. Accountability and legitimacy: society vs donors 4. Conditionality and policy space 5. Aid and state capacity ‘de-building’ 3

ODA: an expanding global complex • Despite shifting trends (eg. aid fatigue in 1990 s), generally significant and systematic increase in number of official donors (around 200 now), NGOs (37, 000? ) and recipient countries (180 for 100 major official donors) • Recently, over 35, 000 annual official aid transactions reported in one year (200 per country) 4

The political economy of aid from the donor’s perspective • Aid as instrument of foreign policy and geopolitical priorities • Aid as economic instrument to further national business interests • Aid as instrument of domestic policy (public opinion, migration, security, etc. ) 5

6

7

Aid effectiveness debates: aid works, it doesn’t, it depends… – Different methodologies – Different time periods / samples – Different policy indicators – Different outcomes – Different explanations: • Destination bias, geopolitics of aid • Perverse macroeconomic effects • Policy environment / ‘bad governance’ • Institutional outcomes and paradoxes 8

Aid volatility in Africa 9

State capacity ‘de-building’: perverse mechanisms • Distortions in government pay structures (per diems, top-ups, etc. ) uneven burden and benefits for civil servants • Distortions in budgeting systems (off-budgets, investment/ recurrent balance) loss of control over budget process • Fragmented and complex aid delivery system inefficient time management Distraction from government programmes and necessary routines loss of capacities to think and articulate long-term strategies • ‘Brain drain’ towards donor agencies and project/implementation units especially in countries with scarce skilled labour growing human resource mobility loss of institutional memory and technical capacities • Reduction in domestic revenue raising capacity through multiplication of efforts to manage aid and debt deepening aid dependence 10

Summary of the growing complexity and irrationality of aid delivery systems General budget support Donor BS review groups Programme aid – SWAPs and their management units ‘Old mechanisms’ - Projects

Aid modalities in Mozambique 12

Conditionality and the loss of policy space • Conditionality alive but frameworks evolve – Expansion of conditions from Washington consensus to ‘Washington confusion’ (Rodrik) – endless shopping list of recommendations – Governance conditionality and move towards ‘selectivity’ – From outcomes to process greater interference through participation in policy-making processes (e. g. PRSPs) • Channels of ‘loss’ of policy space – From imposition to ‘forced consensus’ self-censorship – Donor ‘cartels’ driving explicit and implicit conditionality (dominated by WB/IMF, USA, UK and EC) – Gradual ideological conversion of technocrats and politicians in recipient countries 13

New aid agenda closely linked to ‘good governance’ agenda: the post-Washington consensus • In light of SAP’s failure, focus on institutions ‘getting institutions right’ • Why ‘good governance’? – Fiduciary aspect (need for accountability and transparency) – Alleged positive correlation between ‘good governance’ and development 14

Talk the talk… "We make absolutely clear to countries that transparency and good governance are vital. We are prepared to withhold funding through governments when our standards are not met, as we have done in Malawi. ” Andrew Mitchell, International Development Secretary, January 2012 http: //www. bbc. co. uk/news/uk-politics-16410677 15

Walking the walk? Easterly and Williamson, 2009 16

Sector bias of aid: implications of focus on macro, social policies and governance 17

Sector composition of OECD/DAC aid Distribution of aid by use - 2008 TOTAL DAC EC World Bank France Japan Social and administrative iiiiinfrastructure 39. 2 27. 3 47. 1 29. 7 17. 4 Economic infrastructure 16. 3 24. 1 37. 3 20. 1 36. 3 Production 6. 5 6. 3 14. 8 5. 7 12. 4 Multisector 5. 7 9. 6 0. 8 10. 6 2. 7 Programme assistance 5. 0 18. 7 - 11. 4 4. 4 Sub-total 72. 8 Source: own elaboration from DAC database 86. 0 100. 0 77. 6 73. 2

Aid and ‘good governance’ in practice: Ambiguities and contradictions Donor consensus? – Lack of consensus on what is meant by ‘good governance’ / myriad indicators – Lack of consensus on ‘good enough governance’ – Tension between focus on corruption/politics vs ‘investment climate’ – And many important growthenhancing governance capacities are left out of the picture Contradiction The starlets of DAC donors (Uganda, Mozambique) also characterised by slippage in fundamental aspects of the GG agenda

The political economy of forced/forged consensus: From the Ministry to the IMF/WB and viceversa • Shared sovereign space between BWI and African governments in key institutions: finance ministries and central banks. • More importantly, even greater number of upper-middle-level technocrats have attended training programmes offered or sponsored by BWI and like-minded donors (WB, USAID, DFID) through WBI, AERC, and Anglo-American academic institutions • The WB has complemented this with ambitious support to research capacities and data collection at govt level 20 Source: Van Waeyenberge (2008) http: //www. soas. ac. uk/cdpr/seminars/43473. pdf

Mapping the aid-governance nexus Institutional fragmentation From content to process conditionality Ideology, technologies of policy processes and ‘capacity building’ • State fragmentation • Institutional entanglement btw donors and SSA states • Epistemic communities and shared agendas • Shifting material priorities in allocation of fiscal resources • Logic of ‘aid maximisation’ 21

Agency and rents matter: despite these common problems outcomes may differ Spectrum of ‘government control’ over policy agenda and implemented outcomes Ownership as ‘control’ and not as ‘commitment’ Strongest Bostwana Weakest Ethiopia Rwanda Key issues: structural conditions (economic, geopolitical, etc. ) negotiating political capital Source: Whitfield (2009, p. 331). Ghana, Zambia, Mozambique, Tanzania, Mali

Some conclusions • Aid flows have increasingly problematic governance implications. Question is what kind of governance capacities are created and/or destroyed in the process • Loss of policy space substantial but not complete and as much a product of powerful internal dynamics and social/economic/political changes as a result of external pressures – importance of context • Operational imperatives of aid agencies impair progress towards reforms of aid architecture so change must come particulary from internal/domestic dynamics/agency 23