System Structures
Three structural dynamics operate beneath the African scaling ecosystem's repeated outcomes. Each is self-reinforcing. Each rewards the behaviours that maintain it. None is the failure of any individual actor inside it.
The asymmetry across the three structures is honest. The Capability Trap is the most heavily evidenced through four years of EADC programme research and the broader applied-research base. The Misaligned Incentive Engine has the clearest behavioural signature in the documented patterns of role collapse, programme-output orientation, and evaluation capture. The Capital Architecture Mismatch is the most visible structurally - the offshore incorporation dynamic and the Series B cliff are its surface manifestations. The clustering does not collapse the three structures into a single explanation. It shows that they are connected, that each stabilises the others, and that none can be addressed in isolation.
Three structures generating the outcome. Each structure operates as a self-reinforcing dynamic. Each generates a specific cluster of paradoxes (mapped to the thirteen documented in Paradoxes). The interactions between them - the Capability Trap blocks diagnosis of the other two, the Misaligned Incentive Engine blocks action on diagnosis, the Capital Architecture Mismatch compounds the institutional weakness that traps capability - explain why the equilibrium is stable across multiple correction cycles. Each structure is rational from inside the system. Together they reproduce an equilibrium that serves no one.
Treating symptoms without touching the underlying structures produces temporary relief. The system adapts. New symptoms emerge in adjacent domains. This is the analytical claim the rest of the section develops.
Structure 1: The Capability Trap
The basic dynamic: institutions that lack the capability to solve a problem design ineffective solutions, which fail to generate the evidence that would help those institutions improve, which means the next intervention is also ineffective.
The challenges facing the African scaling ecosystem - market fragmentation, weak infrastructure, absent regulatory frameworks, missing institutional actors - are large, complex, and deeply interdependent. Solving them requires the analytical capability, long-term commitment, and systems understanding that the institutional actors nominally responsible for solving them often do not have. This is a structural condition, not a judgement on individuals operating within it.
The structural foundation for understanding why institutions in this position systematically reproduce ineffective interventions sits in modern development scholarship. Andrews, Pritchett and Woolcock's Building State Capability - and the broader research programme on isomorphic mimicry as institutional failure mode - establishes the canonical mechanism: institutions under pressure to demonstrate capability they do not possess reliably produce activities that look like effective practice without acquiring the function the practice was meant to deliver. The form is reproduced. The function is absent. The reporting metrics confirm the form. The outcomes confirm the absence of function. The two pieces of information do not communicate with each other within the institution's evaluation architecture, which means the cycle continues.
The implication for African ecosystem capability is direct. Governments that cannot understand scaling ecosystems cannot design scaling-specific policy, so they design general SME programmes instead. Innovation agencies that cannot evaluate programme effectiveness cannot learn from failure, so each programme cycle repeats the same approach. The cleanest expression of the dynamic is that the greater the market failure, the less capable the institution nominally responsible for fixing it.
The trap has a particularly self-reinforcing mechanism. It produces a preference for visible, measurable, short-cycle activities over substantive, structural, long-cycle ones. Incubators can be opened. Competitions can be held. Reports can be written. None of these require the capability to understand what scaling ventures actually need. All of them generate the outputs that institutional reporting systems measure. The trap is, in this sense, comfortable for the institutions inside it. The substantive treatment of how this dynamic operates specifically in the acceleration architecture - and the academic-literature evidence on what intensive, capability-building programme design actually requires to produce sustained outcomes - sits in (Stalled) Acceleration. The substantive treatment of how the same dynamic produces the institutional-actor failures across the multilateral, donor, and DFI architecture sits in The Political Economy of the Ecosystem and Institutional Actors.
Breaking the trap requires inserting external capability into the system as embedded actors who transfer knowledge while delivering results, rather than as consultants who deliver a report and leave. It requires measurement systems that track scaling outcomes rather than programme outputs - the information-flows leverage point developed substantively in A Theory of Ecosystem Change. And it requires the institutional discipline to distinguish between activities that produce the appearance of supporting scaling ventures and activities that actually do. The African Scaleup Lab proposition specified in Recommendations is the institutional architecture designed against this structural failure.
Structure 2: The Misaligned Incentive Engine
The basic dynamic: the actors responsible for producing better outcomes for scaling ventures are rewarded for behaviours that are misaligned with those outcomes - and in some cases, actively contrary to them.
The structural foundation for understanding why misaligned incentives persist even when actors recognise the misalignment sits in modern institutional economics. Oliver Williamson's The Economic Institutions of Capitalism - and the broader transaction-cost economics tradition - establishes the foundational principle: institutional arrangements stabilise around the equilibrium where the cost of operating within the arrangement is lower than the cost of changing it. Where reform requires coordinated action by multiple actors whose institutional positions depend on the existing arrangement, the transaction costs of reform exceed the gains any individual actor captures from contributing to it. The arrangement persists not because any actor prefers it but because no actor's individually-rational deviation can produce the collective outcome that would benefit them all.
The implication for the African ecosystem is empirical. Donors are rewarded for programme delivery, not scaling outcomes. Accelerators are rewarded for cohort numbers and demo days, not for the revenue growth and employment creation of their alumni. Evaluators are rewarded for positive reports, not for accurate accounts of what worked. The rational response for each actor is to do what they are rewarded for. The result is a self-reinforcing cycle in which the acceleration model is continuously reproduced regardless of its effectiveness, the funding architecture is continuously reproduced regardless of its outcomes, and the evaluation architecture is continuously reproduced regardless of what it actually measures.
When Multilaterals Compete documents the structural form this takes inside even the most sophisticated ecosystem actors: role collapse - simultaneously funding ventures, running programme support for them, evaluating their own programmes, and competing with the ventures they fund. Role collapse is the rational response to an incentive environment that rewards bundled presence over focused effectiveness. The substantive treatment of how role collapse operates across the multilateral and DFI architecture sits in The Political Economy of the Ecosystem and Institutional Actors.
The same structure operates in the investor layer. The performance evidence on female-founded ventures - documented systematically through the EADC GESI research and threaded through The Gender-Scaling System - is unambiguous on direction. The investment behaviour that the evidence should produce is not occurring. The Misaligned Incentive Engine explains why: investment decisions flow through networks and evaluation criteria built before this evidence existed, with no mechanism to route the new data to the decision-making layer. The system has no self-correction. Where voice is structurally implausible, in Hirschman's framework developed substantively in System Dynamics, the system must be redesigned. Marginal adjustments to programme content do not produce outcomes the incentive architecture is structurally calibrated against.
Breaking the Misaligned Incentive Engine means changing what actors are rewarded for. For donors: outcome-based funding, with independent evaluation commissioned by funders rather than implementers - operationalised in Enabling Conditions 1 and 3 specified in Recommendations. For accelerators: revenue-sharing or equity-based models that align programme success with venture success - the architecture the leading bifurcating ESOs are already migrating toward, treated substantively in (Stalled) Acceleration. For investors: portfolio support quality incorporated into LP reporting and GP assessment frameworks. These are structural changes to who benefits from what - not marginal adjustments to programme design.
Structure 3: The Capital Architecture Mismatch
The basic dynamic: African scaling ventures need a type of capital that the ecosystem does not systematically provide - patient, risk-tolerant, locally denominated, with return expectations calibrated to African operating conditions. What they receive is structurally mismatched with what they need.
The structural foundation for understanding why capital markets fail in the presence of information asymmetries that cannot be cheaply resolved sits in modern financial-economics scholarship. George Akerlof's Quarterly Journal of Economics paper "The Market for 'Lemons'" - the foundational treatment of asymmetric information as a structural source of capital-market failure - establishes the canonical mechanism: where investors cannot cheaply distinguish high-quality assets from low-quality ones, they price all assets against the expected mix, which drives high-quality assets out of the market and stabilises the equilibrium around low-quality assets at depressed prices. The market's failure is rational from inside the information structure. The structure cannot self-correct. Resolving the failure requires changing the information architecture, not the actor decisions within it.
The implication for African scaling-venture capital is direct. African ventures are building in markets with regulatory complexity, currency volatility, infrastructure gaps, and informal sector dynamics that international capital cannot cheaply verify or value. The Africa discount - documented substantively across the publication and disaggregated in Socio-Economic Realities into its currency, contract-enforcement, and political-risk components - is the priced expression of this information asymmetry. Investors price against the mix they cannot disaggregate. Founders accept terms they would not accept under symmetric information. The architecture stabilises around an equilibrium where the capital structure that would actually fit African scaling ventures - patient equity at growth stage, locally denominated debt with covenants calibrated to African cash-flow patterns, instruments designed for the physical-digital hybrid models that characterise African scaling - barely exists at scale. What exists is short-tenure, dollar-denominated, equity-structured capital with growth expectations calibrated to other markets.
The mismatch is most structurally visible in offshore incorporation. Offshoring African Startups documents the mechanism in operational detail; the substantive treatment of how the offshore-incorporation requirement operates as a structural barrier to domestic capital mobilisation sits in Political & Regulatory Barriers, Feedback Loops Loop 5, and The Political Economy of the Ecosystem. The pattern matters here for what it reveals about the architecture: founders who successfully scale become less embedded in the local ecosystem as investors, mentors, and advocates because their institutional home is now international. The flywheel that should power the next generation of founders leaks at precisely the point where it should be strongest. This is the Akerlof asymmetry made empirical: the high-quality assets the architecture would most benefit from retaining are precisely the ones the architecture's information failure drives offshore.
The domestic capital dimension of the same structure runs in parallel. The informal capital mechanisms ordinary Africans use - stokvels in South Africa, chamas in Kenya, tontines in Francophone West Africa, equub in Ethiopia - work because they operate with high trust, low transaction costs, and strong local accountability. The formal investment infrastructure through which wealthy Africans participate in capital markets is oriented in the opposite direction - toward international structures, dollar returns, and distant assets. Domestic institutional capital - pension funds, sovereign wealth funds, insurance companies - is similarly misaligned, currently underinvested in African scaling ventures relative to its long-run interest in domestic economic growth. The substantive treatment of why pension regulatory architecture and DFI mandates produce this misalignment - and the specific reforms that would address it - sits across the Capital Systems chapter and operationalised in Enabling Conditions 6, 7, 8, and 9 specified in Recommendations.
Breaking the Capital Architecture Mismatch requires structural diversification: more revenue-based financing, more local currency debt, more patient equity, and genuine DFI risk appetite at the growth stage. It requires regulatory reform that creates enabling rather than obstructing conditions for cross-border venture operation. It requires deliberately designed diaspora investment channels that work with actual investor incentives. None of these is exotic. All are within existing institutional authority. The reason they have not been implemented is the political economy of the existing architecture - which is the Misaligned Incentive Engine operating on the Capital Architecture Mismatch.
How the three structures interact
The three structures do not operate independently. Their interaction is the reason the equilibrium is stable across multiple correction cycles.
The Capability Trap blocks diagnosis of the other two. Institutions that lack the analytical capability to understand scaling ecosystems cannot diagnose the Misaligned Incentive Engine in their own programme architecture or the Capital Architecture Mismatch in their own funding instruments. Governments that cannot evaluate scaling outcomes cannot identify which regulatory reforms would dissolve the offshore-incorporation pressure. Innovation agencies that cannot measure programme effectiveness cannot identify the incentive misalignments that perpetuate ineffective acceleration. The first interaction is informational: the trap stabilises by preventing the diagnosis that would shift the other two.
The Misaligned Incentive Engine blocks action on diagnosis. Even where actors have the capability to diagnose structural problems, they are not rewarded for acting on the diagnosis. Donors who understand that acceleration does not reliably produce scaling outcomes continue to fund acceleration because that is what their reporting systems reward. The donor funding contraction of 2025–26 - documented substantively in Lessons from the Correction - does not resolve this. It removes the resource without changing the incentive logic. The second interaction is behavioural: the engine stabilises by ensuring that even diagnosed structures cannot be acted on within the existing reward architecture.
The Capital Architecture Mismatch compounds the institutional weakness that traps capability. Where domestic institutions cannot mobilise the capital that would build the institutional capability they lack, they remain dependent on external capital whose mandates reproduce the conditions for the trap. The third interaction is resource-flow: the mismatch deprives institutional actors of the financial autonomy that would allow them to escape the dependency that traps them.
The result is a system that is remarkably stable. The same dynamics generate the same outcomes reliably, year after year, regardless of which individuals are operating within them. The outcomes themselves are poor and worsening on gender equity, geographic concentration, and the Series B cliff. The diagnosis is named here. The mechanism through which each structure produces its specific paradoxes is treated substantively in Paradoxes. The self-reinforcing dynamics that make the equilibrium stable across correction cycles are treated substantively in Feedback Loops. What it would take to shift the structures is the subject of the publication's Way Forward.