After five years of analysing what institutions are and why they fail, the question that follows is: what institutions would you build, given the chance?
From Diagnosis to Design
Institutional analysis is incomplete if it can describe failure but cannot say what improvement requires. The constructive task is not to imagine institutions without conflict, error or waste. Such institutions do not exist. It is to build institutions in which failure becomes visible, responsibility can be located, correction is possible and the people who depend on the institution possess more than a ceremonial voice.
The institutions I would build would share four properties: genuine accountability to the populations they serve rather than nominal accountability to distant overseers; incentives aligned with institutional purpose rather than institutional survival; transparent performance information that permits honest judgment; and adaptive governance capable of learning from feedback instead of suppressing it.
These properties are not independent virtues. Accountability without information produces grievance without leverage. Transparency without consequences produces disclosure without correction. Incentives attached to the wrong measures turn competent people toward the wrong work. Adaptation without a stable purpose becomes drift. Institutional design must therefore join the four into a single operating system.
What the Four Properties Require
Accountability requires more than identifying an official who can be blamed. It requires a chain connecting authority, resources, expected performance, information and consequences. The World Bank’s World Development Report 2004 described service delivery through relationships among citizens, policymakers, provider organisations and frontline workers, and defined accountability in terms that included delegation, financing, performance information and enforcement. The practical lesson is simple: every mandate must be matched with the means to perform it, every promise with an observable standard and every standard with some power to reward, correct or remove.
Aligned incentives require the institution to reward the achievement of its purpose rather than the expansion of its budget, staff, jurisdiction or public visibility. Survival is necessary, but it cannot become the mission. If promotion follows activity rather than results, the institution will manufacture activity. If funding depends on concealing failure, failure will be concealed. The measurements, career paths and budget rules of an institution reveal its real purpose more reliably than its charter does.
Transparency means that performance information must be timely, intelligible and usable by the people expected to act on it. Publishing a report that no citizen can interpret, or releasing information after the opportunity for correction has passed, satisfies a disclosure rule without creating accountability. Transparency matters only when it reduces the information advantage of those operating the institution and gives others a basis for judgment.
Adaptive governance requires protected mechanisms for admitting error, testing alternatives, ending programmes that do not work and expanding those that do. Feedback is not institutional learning merely because it has been collected. It becomes learning when it can alter budgets, procedures, personnel or strategy. An institution that solicits criticism but cannot change in response has built a listening ritual, not an adaptive system.
Where I Would Build
In the African contexts to which I have repeatedly returned, I would begin in three fields: local governance, financial infrastructure and knowledge infrastructure. These are not claims that one institutional design can fit a continent. They are priorities whose forms must vary by country, city, district and community.
The first is accountable local governance: institutions capable of translating national authority and public resources into services that people can observe where they live. Their budgets, responsibilities and service standards should be legible locally. Citizens should be able to compare promises with delivery, report failure through channels that reach someone with authority and see what follows from the report. Local participation without resources is theatre; decentralised resources without scrutiny merely relocate the opportunity for capture. The institution must combine proximity with enforceable responsibility.
The second is financial infrastructure connecting informal economic activity to productive investment. The International Labour Organization’s estimates based on 2016 data placed informal employment at 85.8 per cent of total employment in Africa. The World Bank’s regional Global Findex analysis also records the continuing importance of borrowing from friends, family and community savings groups in Sub-Saharan Africa. The design problem is therefore not simply how to force economic life into inherited formal categories. It is how to build trustworthy payment systems, savings instruments, credit records, risk-sharing mechanisms and investment channels that can recognise irregular incomes and small transactions while giving viable enterprises a path toward larger and longer-term capital.
The third is knowledge infrastructure: research institutions, policy-analysis capacity, statistical systems, archives and institutional memory. The African Union’s Strategy for the Harmonization of Statistics in Africa calls for sustainable statistical capacity, evidence-based decision-making and methods adjusted to African realities. The purpose of such infrastructure is not to reject knowledge produced elsewhere. It is to ensure that imported models must survive contact with local evidence, and that lessons generated within African institutions can accumulate instead of disappearing whenever personnel or governments change.
These three fields also depend on one another. Local government cannot be accountable without reliable information. Financial institutions cannot assess opportunity without records and knowledge. Research institutions cannot remain relevant if they are detached from the operational problems of governments, firms and communities. The objective is not to construct isolated organisations, but to build an institutional ecology in which information, authority, capital and correction can move.
The Test of Durability
The final test is not whether an institution survives. Institutions can preserve themselves long after they have ceased to serve their purpose. The test is whether the institution can remain useful while conditions change: whether it can retain its mandate, expose its own failures and revise its methods without treating every demand for correction as a threat.
The institutions worth building are the ones that can still be genuinely serving their populations in fifty years because accountability, incentive alignment, transparency and adaptive governance were built into their operating systems from the beginning. Durability is not longevity alone. It is the preserved capacity to correct course without abandoning purpose.
Discussion