Aid effectiveness was a question about paperwork

The administrative diagnosis

By the early 2000s, development finance had accumulated several decades of procedural sediment. A single country ministry might receive dozens of donor missions in a year, each with its own reporting format, its own procurement rules, its own parallel management unit staffed with donor-funded consultants sitting outside the civil service. The cumulative transaction cost was not hypothetical: it fell on the recipient government's own staff, who spent working weeks filling in reports whose information they already held in different formats for different audiences. The money arriving was real; the machinery consuming a portion of its value was also real.

The Paris Declaration on Aid Effectiveness, adopted in March 2005 under OECD auspices, was a collective acknowledgement of that diagnosis. More than a hundred governments and agencies signed it — donor countries, partner countries, multilateral institutions including the World Bank and the IMF. What they agreed to was not a new financial commitment but a set of working practices, organised under five principles: ownership, alignment, harmonisation, managing for results and mutual accountability. Most principles had at least one indicator attached to them, and each indicator had a target to be reached by 2010.

Alignment, harmonisation, untying

Alignment meant that donors would use the partner country's own systems — its budget processes, its procurement rules, its financial management framework — rather than erecting a parallel structure beside them. In practice, this was harder than it sounded. Donors had their own auditing obligations, their own legal requirements around fiduciary risk. Using a country system that their own assessments rated as weak exposed them to political accountability at home. The indicator for alignment tracked what proportion of aid flowed through government financial management systems; the 2011 survey published by the OECD found that progress was limited, with the share moving only modestly across most partner countries.

Harmonisation addressed the specific problem of duplication. Where multiple donors funded similar activities, the declaration asked them to coordinate — pooling funds, sharing analytical work, conducting joint missions, producing common reporting templates. The Rome Declaration of 2003, a precursor meeting, had already flagged this; Paris gave it a numerical target. Measuring harmonisation meant counting the proportion of aid delivered through coordinated programmes, which required defining what "coordinated" meant, a definitional question the monitoring surveys handled differently across rounds.

Untied aid was a distinct but related strand. Tying aid — requiring that goods and services purchased with it be sourced from the donor country — reduced its purchasing power in recipient economies, sometimes substantially. Estimates of the efficiency loss from tying varied, but the direction was not disputed. The Paris Declaration committed signatories to progress on untying, building on a 2001 OECD recommendation on untying aid to the least developed countries. Progress was reported through the OECD's Creditor Reporting System, which tracked whether aid was formally tied, partially tied or untied — a classification that depended on donors' own self-reporting.

What the monitoring showed

The monitoring architecture for Paris was itself a substantial piece of machinery. The OECD's monitoring surveys, conducted in 2006, 2008 and 2011, collected data from partner countries and donors on all twelve indicators. The 2011 survey — the final one before the agenda transitioned to the Busan outcome — covered 78 partner countries and produced country-level scorecards alongside aggregate figures. Only one of the thirteen targets set for 2010 was met in full.

That finding requires careful reading. The targets had been set without a rigorous baseline: the 2005 survey established the starting point only retrospectively, meaning the ambition of the 2010 targets was calibrated against data that was itself being assembled for the first time. Some indicators proved difficult to measure consistently. Mutual accountability, the fifth principle, had no quantitative indicator at all until the Accra Agenda for Action of 2008 sharpened the framework. Managing for results — the expectation that both donors and partners would orient decisions around outcome data rather than input compliance — sat uneasily with the indicator-based monitoring of the declaration itself, which measured process rather than development outcomes.

Accra and Busan

The Accra Agenda for Action, agreed in Ghana in 2008, acknowledged that implementation had been slower than expected and added emphasis on country ownership and transparency. It introduced stronger language on aid predictability — whether disbursements matched commitments and schedules — because unpredictable flows disrupted budget planning in ways that offset any gains from harmonisation. The High Level Forum in Busan in 2011 produced the Busan Partnership for Effective Development Co-operation, which broadened the framework to include new actors: emerging economy donors, private sector participants, civil society organisations. It also endorsed the International Aid Transparency Initiative's standard as a transparency mechanism, and the Global Partnership for Effective Development Co-operation, established at Busan, took over the monitoring function from the OECD DAC.

What changed between Paris and Busan was partly the cast of participants and partly the vocabulary. "Aid" became "development co-operation" to accommodate flows that did not fit the OECD DAC's official definition — south-south cooperation, export credits, blended finance instruments. The monitoring indicators were revised, reducing the emphasis on the procedural plumbing of alignment and harmonisation and increasing attention to transparency and results. This shift meant that the Paris series and the post-Busan series are not straightforwardly comparable, which complicates any long-run assessment of whether the agenda worked.

The limits of the instrument

What the Paris framework could not reach was the political economy beneath the procedural surface. A donor government that harmonised its reporting formats with others was still making decisions about what sectors to fund, which countries to prioritise and what conditions to attach — decisions driven by foreign policy, domestic constituencies and institutional incentives that a declaration of principles did not alter. Alignment required trusting partner country systems, which required a judgment about fiduciary risk that each donor institution made internally against its own board. The paperwork could be reformed without the underlying pattern of incentives changing much.

The 2011 evaluation by the OECD and the World Bank found genuine improvements in some areas — mutual assessment reviews became more common, joint missions increased in a number of countries — alongside persistent gaps in alignment and slow progress on untying. What the monitoring surveys documented, accurately, was the state of the process. Whether a more efficiently administered aid programme produced better development outcomes than the counterfactual was a question the framework was not designed to answer, and the data it generated were not the data needed to try.