Who owns the plan, and who agreed to stop asking

The architecture of ownership

The 2005 Paris Declaration on Aid Effectiveness, signed by more than a hundred governments and donor agencies under OECD auspices, did not invent the concept of country ownership. The idea had circulated through the development machinery for years before Paris, surfacing in the Poverty Reduction Strategy Paper process the World Bank and IMF introduced at the turn of the millennium and appearing, more quietly, in the Rome Declaration on Harmonisation of 2003. What Paris did was give ownership operational content: a recipient government's own national development strategy was to be the reference point for aid, and donors were to align their programmes behind it rather than designing parallel structures. The declaration's first principle and its second were, in this sense, a single proposition described from two sides.

The practical consequence was specific. Alignment meant that a donor financing a health ministry should use the ministry's own planning documents, its procurement systems, its financial management procedures, and its reporting formats — not import a set of parallel requirements the ministry then had to satisfy alongside its own work. Harmonisation extended this further: where multiple donors were active in the same sector, they were to coordinate among themselves so the ministry faced one set of procedures rather than many. Both commitments required donors to relinquish something they had institutional reasons to keep — direct visibility over how their money was spent, and the reporting lines that provided it.

The distance between principle and practice showed up almost immediately in the Paris Declaration monitoring surveys, which the OECD Development Assistance Committee ran in 2006 and 2008 against twelve indicators the declaration had itself defined. The indicator for alignment with country systems — measuring what share of aid flowed through recipient-government procurement and financial management — produced results that were well below the targets the signatories had set for 2010. Donors continued to operate project implementation units, to require audits by firms of their own choosing, and to demand reporting in formats their own headquarters could process. The monitoring data recorded these gaps without attributing blame: the surveys measured what happened, not why.

What donors were agreeing to give up

The parallel reporting problem had a structural cause that no declaration could dissolve by assertion. A donor disbursing public funds to a foreign government remained accountable to its own legislature and auditor-general. When a country's public financial management systems were assessed as weak — as they were in much of sub-Saharan Africa according to the World Bank's Country Policy and Institutional Assessment scores — donors faced a genuine tension: aligning with those systems meant accepting a level of fiduciary risk their own governments might not sanction. The Paris Declaration acknowledged this by building in conditionality of its own kind, asking recipient countries to demonstrate progress on public financial management as a precondition for deeper alignment. In effect, ownership was offered as a destination, not a starting point.

This conditionality sat awkwardly alongside the declaration's rhetoric. If a government had to earn the right to lead its own strategy, the principle of ownership was already qualified. The Accra Agenda for Action, agreed in Ghana in 2008, attempted to address this by pushing donors to commit more explicitly to using country systems even where those systems were imperfect, and to build capacity rather than substitute for it. Accra also called for more predictable, multi-year aid commitments — a precondition for any government planning process that deserved the name. Whether those commitments were honoured is a question the monitoring data answered only partially: the 2011 survey, the last major one conducted under the Paris framework, showed uneven progress and some reversals.

The Busan Partnership for Effective Development Co-operation, concluded in South Korea in 2011, reframed the language again. "Ownership" became "country-led development", partly to accommodate a wider set of actors — emerging economies, private finance, civil society — who had not been signatories to Paris and whose engagement the framework needed. The Busan text was less binding in form than Paris had been, a reflection of the political difficulty of securing agreement across a much more varied group. The Busan outcome document retained the language of alignment and mutual accountability, but the monitoring architecture it inherited was contested: some of the largest emerging donors did not accept the Paris indicators as legitimate measures of their behaviour.

Where the uncertainty accumulates

Measuring ownership is intrinsically harder than measuring, say, the share of aid that is untied. Untied aid has a definition that, however disputed at the margin, produces a countable number. Ownership does not reduce to a single indicator. The Paris monitoring framework approximated it through proxies: whether a country had an operational national development strategy, whether donors aligned their programmes to it, whether joint reviews replaced separate ones. Each proxy captured something and missed something else. A government could produce a poverty reduction strategy that satisfied the formal test while the real decisions were made elsewhere; donors could tick the alignment box while maintaining informal parallel requirements. The gap between the indicator and the underlying concept was, in this case, unusually wide.

UNICEF and other agencies operating in social sectors documented a specific version of the problem: sector-level planning processes that nominally drew on government education or health strategies but in practice shaped those strategies from outside, through the conditionalities attached to sector-wide approaches. The strategy was government-owned in provenance; the content had been negotiated. Whether that counts as ownership depends on a theory of the concept that no monitoring framework could settle.

The statistical offices that produce the underlying figures — national statistics offices rather than international agencies — were largely peripheral to the ownership debate, which played out in planning and finance ministries. But they bore a related pressure: donors wanted data that met their own reporting standards, which sometimes meant parallel data collection exercises running alongside national household surveys, consuming the same thin statistical capacity. Here too the alignment principle and the parallel-reporting habit were in direct conflict, and here too the habit generally won.

What the Paris–Accra–Busan sequence established is a documented record of what governments and donors agreed to do, what indicators they chose to measure it, and what the surveys showed. The record is not one of failure — some alignment indicators moved in the right direction, and the five principles became the reference point for subsequent aid architecture debates. It is a record of a principle — ownership — that was structurally in tension with the accountability obligations of the institutions asked to honour it, and of an indicator framework that made that tension visible without resolving it.