Infrastructure is only transformational when it becomes productive capacity

The strongest infrastructure programmes are not necessarily the largest. They are the programmes that remove constraints that suppress activity across a production network and that sequence physical investment with the institutional, regulatory, financial and operational changes required to use the asset.

This matters because infrastructure decisions are often made through separate administrative silos. Transport authorities select roads. Energy agencies plan generation. ports authorities expand terminals. digital ministries fund networks. finance ministries manage fiscal exposure. regulators set tariffs and standards. investment agencies court investors. Yet the economic return emerges from the interaction of those systems. A manufacturer does not experience infrastructure as separate ministries. It experiences the reliability of power, the time required to move inputs, the predictability of customs, access to finance, digital connectivity, standards compliance and the quality of public services as one operating environment.

The World Bank's recent infrastructure work reinforces this shift from expenditure to outcomes. Its analysis connects infrastructure stocks, costs and financing conditions with growth, jobs and productivity while highlighting investment efficiency, existing capital stocks, construction costs, procurement and market structure as important determinants of results. The implication is straightforward: increasing the volume of spending does not automatically increase the development return on spending.

The African Development Bank's industrialisation evidence adds a second layer. Africa's manufacturing base has expanded, but the continent remains a small share of global manufacturing output and exports. Infrastructure is not the sole explanation. Productive transformation also depends on skills, finance, standards, technology, market access, competition and institutional capability. The relevant question is therefore not whether infrastructure causes industrialisation by itself, but whether infrastructure is being designed as part of the production system required for firms to become more competitive.

TAY's assessment is that infrastructure policy should move from an asset-centred pipeline to a constraint-centred pipeline. The starting point should be the economic problem: what prevents firms, workers, public agencies or markets from producing, trading or investing at greater scale? The next step is to identify the combination of physical infrastructure, policy, institutions and capital that can remove that constraint at acceptable cost and risk.

The conventional infrastructure pipeline begins with an asset. A government identifies a road, railway, power plant, port, industrial park, water system or digital network and then asks how to fund it. A productive-capacity approach begins earlier. It asks what economic capability is missing and whether infrastructure is actually part of the solution.

This distinction is material. Suppose an agricultural region has low export volumes. A conventional response might be to construct a major road. A systems approach would first identify the full chain: farm productivity, aggregation, storage, cold chain, feeder roads, electricity, standards, border clearance, port logistics, working capital, insurance and market information. If the binding constraint is post-harvest loss or lack of cold storage, a new trunk road may not deliver the expected export increase. If the constraint is a missing corridor link that makes existing production commercially inaccessible, the road may have very high value.

The same reasoning applies to industrial power. Installed megawatts are a weak proxy for productive electricity. An industrial firm needs predictable voltage, connection capacity, maintenance, tariffs it can plan around and a utility capable of restoring service. A generation project may therefore be necessary without being sufficient. The investment case depends on the network around it.

The principle is general: infrastructure creates productive capacity when it changes the operating conditions of economic activity. That means the analytical unit should often be the production network rather than the physical asset.

Infrastructure failure is often visible at the asset but caused elsewhere in the system. A congested port may reflect customs procedures, inadequate hinterland transport, documentation requirements, yard management or poor scheduling rather than berth capacity alone. A poorly used industrial zone may reflect land administration, power reliability, skills, market access or financing rather than the absence of factory buildings. A digital public platform may underperform because agencies do not share standards or data even though the software itself works.

This is why TAY uses a binding-constraint test before treating an infrastructure gap as the primary problem.

The test asks five questions:

1. What economic activity is being suppressed?
2. Where in the production or service chain does the constraint actually sit?
3. Which institution has authority to change it?
4. What combination of infrastructure, policy and capital can change the constraint?
5. What observable outcome would demonstrate that the constraint has been removed?

The purpose is not to minimise infrastructure investment. It is to improve its targeting. A relatively small intervention that removes a high-cost bottleneck can generate more additional economic activity than a much larger project that duplicates existing capacity.

This also improves public accountability. When the desired outcome is defined before construction, project sponsors can identify complementary reforms, assign institutional responsibilities and measure whether the investment is producing the expected service. Without that chain, completion becomes the easiest metric to report because it is the easiest metric to observe.

TAY's core infrastructure transmission mechanism is:

asset → service reliability → transaction cost and risk → investment and utilisation → production and trade → jobs, fiscal capacity and resilience

Each link matters.

Asset to service

The physical asset must provide a usable service at the required quality. A road must remain passable. A power system must deliver dependable electricity. A port must process cargo efficiently. A water system must supply users consistently. A digital network must provide reliable and secure connectivity.

Service to cost and risk

Reliable service changes the economics faced by firms and households. It can reduce travel time, inventory requirements, spoilage, downtime, uncertainty, insurance exposure and working-capital needs. Reliability is therefore an economic variable, not simply an engineering metric.

Cost and risk to investment

When firms can forecast operating conditions, they are more able to invest. Predictability affects plant utilisation, location decisions, supplier relationships, inventory strategies and the willingness of lenders and equity investors to commit capital. Infrastructure can therefore have a mobilisation effect that is larger than the direct value of the asset itself.

Investment to production and trade

The ultimate test is whether firms expand productive activity, whether supply chains become deeper, and whether markets become more accessible. This can mean increased manufacturing, agro-processing, mining services, logistics, tourism, digital services or other activities depending on the local economic structure.

Production to public outcomes

Higher productive activity can broaden employment, exports, tax revenue and household income. It can also improve resilience by diversifying suppliers and markets. But these benefits are not automatic. They depend on whether complementary constraints are addressed and whether the state can capture sufficient fiscal value to maintain the infrastructure and services.

Industrialisation requires firms to coordinate many inputs at once. A factory needs electricity, transport, water, communications, land, workers, suppliers, standards, finance and access to customers. A country can possess each component in isolation and still lack a competitive production environment.

This is why industrial corridors and economic zones should be evaluated by firm-level outcomes rather than construction completion. An industrial park with roads and buildings but unreliable power and weak tenant services is an asset portfolio, not yet a productive cluster. A corridor with a major highway but inefficient borders and limited logistics services may move vehicles without creating a deeper production network.

The AfDB's industrialisation evidence is relevant here. Africa's manufacturing value added has grown, but the continent remains a small participant in global manufacturing. Closing that gap requires more than physical infrastructure. It requires coordinated productive capabilities: firms that can meet standards, workers with relevant skills, financiers willing to fund expansion, logistics systems that connect suppliers and customers, and institutions that make the rules predictable.

Infrastructure policy should therefore be integrated with industrial policy. When a government identifies a strategic production opportunity, infrastructure planning should ask what service conditions are required for firms to compete. Conversely, infrastructure planning should identify the production systems likely to use new capacity rather than assuming demand will appear after construction.

This does not mean governments should pick winners mechanically. It means investment should be connected to observable economic demand, market opportunities and credible capability-building pathways.

A corridor is often reported through kilometres of road or rail and traffic volumes. Those are useful operational indicators, but they are not the full economic proposition.

A productive corridor connects farms to processors, mines to industrial services, manufacturers to suppliers, ports to inland markets and firms to regional customers. Its value grows when complementary systems move with it: electricity, storage, customs, standards, digital documentation, payments, insurance, trade finance and industrial land.

This creates a practical distinction between movement infrastructure and economic connectivity. Movement infrastructure lowers physical distance. Economic connectivity lowers the total friction of participating in a market.

The second is the stronger development objective.

For corridor planning, the relevant outcome indicators may therefore include border clearance time, logistics cost per unit, port dwell time, inventory days, freight reliability, industrial occupancy, supplier participation, export volumes, processing capacity and private investment mobilised along the corridor.

A corridor that carries more trucks but does not reduce the cost of production may be improving traffic without transforming the economy. A corridor that enables new processing plants, suppliers and regional trade can create value well beyond the road itself.

Energy illustrates the distinction particularly clearly. Installed generation is a necessary indicator of system capacity, but firms purchase reliable electricity service rather than megawatts on a spreadsheet.

A power system can therefore underperform despite substantial investment if transmission constraints, distribution losses, connection delays, maintenance failures, tariff distortions, utility arrears or weak governance prevent capacity from reaching users. Industrial demand may remain suppressed because firms cannot trust the service or cannot justify connection costs.

The infrastructure investment case should consequently track a chain such as generation capacity → transmission availability → distribution reliability → industrial connection → electricity cost and quality → firm utilisation and investment.

This also changes financing logic. A generation asset with predictable cash flows may support commercial financing, while transmission, rural access or institutional reforms may require different instruments. A guarantee may address a defined payment risk. Concessional finance may support a public-value component. Project preparation may resolve technical and commercial uncertainty. The financing structure should follow the economic function rather than forcing every component into one model.

The same principle extends to digital systems. Fibre, cloud capacity, data centres, digital identity, interoperable payments and secure data exchange increasingly determine how firms and public agencies operate.

But connectivity alone is not productive digital infrastructure. The system must be affordable, interoperable, trusted and usable. A government platform that cannot exchange data with another agency creates administrative friction even if both platforms are technically functional. A payment system that excludes smaller providers limits network effects. A data centre without sufficient power reliability, connectivity and demand can become an underutilised asset.

Digital infrastructure therefore reinforces the broader TAY thesis: infrastructure value is produced by the system around the asset.

The institution with the mandate often matters as much as the institution with the money.

Infrastructure projects cross mandates. Finance ministries manage fiscal exposure. sector ministries define policy. regulators set economic rules. utilities operate services. customs controls borders. local governments manage land and local services. investment agencies engage firms. development banks provide finance and technical support. Private operators may build and manage assets.

A project can therefore fail without any single institution being incompetent. The failure can occur between mandates. One agency optimises construction, another optimises fiscal cost, another protects regulatory stability, and another focuses on attracting investment. If there is no mechanism that joins these objectives around an economic outcome, the project can be individually rational and collectively weak.

Institutional mapping should therefore be part of infrastructure appraisal. For each material constraint, decision-makers should identify who has authority, who provides capital, who bears risk, who operates the service, who regulates it and who benefits from the outcome.

This is especially important for public-private partnerships. PPIAF's project-preparation work emphasises the importance of pipeline identification, screening, preparation, credible information, appropriate risk allocation and capable institutions. Private capital cannot compensate indefinitely for weak project definition or unresolved public-sector responsibilities.

Infrastructure finance is often discussed as a question of volume: how much money is available and how large the financing gap is. The more useful question is whether the available instrument matches the project's risk, cash-flow profile and public value.

Public budget resources can be appropriate for assets with broad benefits and weak direct cash flows. User charges can support commercially viable services where affordability and regulation permit. Long-term debt can finance assets with predictable revenues. Equity can absorb construction and operating risk. Guarantees can address specific credit or payment risks. Development finance can support project preparation, institutional reform, risk allocation or local-currency structures. Blended finance can be justified where concessional resources address a defined market failure and crowd in additional investment.

The discipline is to identify the constraint first and the instrument second.

A financing package that merely makes an uneconomic project affordable does not create transformation. A financing package that resolves a specific bankability constraint while improving the underlying economics can be catalytic.

This is also where project preparation becomes strategic. Investors require credible technical information, demand assumptions, contracts, risk allocation, regulatory clarity and capable counterparties. Spending resources early to resolve those issues can increase the probability that later private or institutional capital is deployed efficiently.

An asset that cannot be maintained is not a completed infrastructure investment in economic terms.

Maintenance expenditure, operating costs, debt service, tariff policy, subsidy obligations and contingent liabilities all determine whether service quality persists. Governments can therefore create an infrastructure stock that appears large while the effective service stock deteriorates.

The fiscal test should be integrated from the beginning. Project appraisal should examine not only construction cost but lifecycle cost, revenue assumptions, foreign-exchange exposure, debt-service implications, subsidy requirements and the government's capacity to meet maintenance obligations.

This is especially important where infrastructure generates benefits that are public or indirect. The state may reasonably fund an asset whose economic benefits exceed its direct revenues, but the fiscal architecture must still make the recurring service sustainable.

The question is not simply whether a government can finance construction. It is whether it can sustain the service without destabilising the wider fiscal system.

Governments should move toward constraint-based infrastructure pipelines.

Before approving a major asset, the responsible authority should be able to state the economic constraint being addressed, the users who are affected, the complementary systems required, the institutions responsible for those systems, the financing structure, the principal risks and the outcome indicators that will be monitored after completion.

Project selection should also distinguish between strategic infrastructure and politically visible infrastructure. A smaller project that removes a critical bottleneck may deserve priority over a larger project with weaker additionality. That is a technical assessment, not a judgement about the political attractiveness of projects.

Governments should also treat maintenance, regulation and operating capability as part of the investment rather than post-construction administration.

For SOEs, utilities, ports, rail operators and other state-linked infrastructure entities, performance should be measured through service quality and economic utilisation as well as financial statements.

Relevant indicators can include uptime, connection time, turnaround time, throughput, asset utilisation, maintenance backlog, operating cost, collection efficiency, customer reliability and downstream economic activity.

The operator is not merely the custodian of an asset. It is part of the mechanism through which public capital becomes an economic service.

This creates a direct connection between corporate governance and development impact. A well-financed asset can still underperform if operating incentives, procurement, maintenance systems or management accountability are weak.

Corporate decision-makers should assess infrastructure as part of the operating environment rather than as a binary question of whether an asset exists.

Investment committees should ask whether power, logistics, customs, standards, digital connectivity, land, skills and supplier networks are reliable enough for the proposed business model. Where a project depends on a public infrastructure improvement, the timing and credibility of that improvement should be treated as an investment risk.

This creates opportunities for firms that can identify infrastructure-linked bottlenecks and build services around them. Logistics, warehousing, energy services, industrial utilities, digital infrastructure, testing and certification, maintenance and specialised project services can become productive complements to public infrastructure.

DFIs should distinguish between financing assets and financing the conditions under which assets become productive.

In some cases, the highest-value intervention may be project preparation rather than a larger loan. In others, it may be a guarantee, local-currency structure, transaction advisory package, institutional reform programme or complementary investment that makes the core asset viable.

DFIs can also strengthen discipline by tying support to measurable outcomes and by requiring credible operating and maintenance arrangements. Mobilisation should be assessed not only by the amount of private capital attracted but by whether the resulting investment improves the underlying economic system.

The public-value case for concessional capital is strongest when it addresses a constraint that commercial capital cannot efficiently absorb and when the intervention creates a credible pathway toward sustainable financing.

A serious infrastructure programme should establish an outcome dashboard before construction begins.

Service reliability: uptime, outage frequency, travel-time reliability, connection delays, water availability or network quality.

Transaction cost: logistics cost, border clearance time, port dwell time, inventory days, administrative processing time and digital transaction cost.

Investment mobilisation: additional private investment, new connections, project financial close, industrial occupancy and supplier investment.

Productive capacity: production volumes, capacity utilisation, processing throughput, manufacturing value added, agricultural value captured or service-sector output.

Market access: export volumes, number of markets served, regional trade flows, supplier participation and corridor utilisation.

Fiscal sustainability: operating cost recovery where appropriate, maintenance funding, debt-service exposure, contingent liabilities and lifecycle cost performance.

Distribution and resilience: access for underserved areas, employment effects, supplier participation, climate resilience and continuity of essential services.

The exact indicators should vary by sector. The principle is constant: measure the economic service created, not only the asset constructed.

The infrastructure thesis has limits. Infrastructure does not automatically create industrialisation. Human capital, competition, technology, institutions, macroeconomic stability, access to finance and market demand can be binding constraints independently or simultaneously.

There is also an attribution problem. Infrastructure projects operate inside changing economies. A rise in exports may reflect exchange-rate movements, commodity prices, trade agreements, firm-level investment or policy reforms occurring alongside the infrastructure project. Outcome indicators should therefore be interpreted alongside sector and macroeconomic evidence rather than treated as simple proof of causality.

There is a second risk: the productive-capacity framework can become too broad. If every development constraint is included in an infrastructure appraisal, accountability becomes diffuse. TAY's response is to maintain a binding-constraint discipline. The analysis should identify which constraint the infrastructure intervention is intended to remove and which complementary factors are material to that mechanism.

A third risk is financing substitution. The availability of concessional or blended finance can create incentives to finance projects because funding is available rather than because the economic case is strong. Capital architecture should therefore follow the economic function, not the other way around.

Finally, infrastructure benefits can take years to materialise. Early indicators should focus on service quality and utilisation while longer-term indicators track productivity, investment, trade and fiscal effects.

TAY's recommended sequence is deliberately practical.

First: define the economic constraint. Identify the production, trade, service or public-capability problem that is being suppressed.

Second: map the system. Identify the asset, service, institutions, complementary infrastructure, firms, users and markets involved.

Third: identify the binding point. Determine which constraint is materially limiting the system and whether infrastructure is actually the most relevant intervention.

Fourth: design the transmission mechanism. Specify how the investment is expected to change service quality, cost, risk, investment, utilisation and output.

Fifth: assign institutional responsibility. Identify who builds, operates, regulates, finances, maintains and monitors each critical component.

Sixth: select the capital architecture. Match public finance, debt, equity, guarantees, concessional resources or blended structures to the risk and cash-flow profile.

Seventh: establish outcomes before construction. Define the indicators that will determine whether the intervention is producing its intended economic effect.

Eighth: stress-test the system. Test demand, operating capacity, maintenance, fiscal exposure, foreign-exchange risk, regulation, complementary infrastructure and institutional coordination.

Ninth: monitor utilisation, not just completion. Continue the investment case into operations and adjust where the expected transmission mechanism is not materialising.

This approach turns infrastructure from a list of projects into an economic transformation pipeline.

Infrastructure should be treated as a system for producing economic services, not as a stock of completed physical assets.

The central decision question is therefore not “What infrastructure should be built?” It is “Which constraint is suppressing productive capacity, and what combination of infrastructure, institutions, policy and capital will remove it?”

That shift has consequences across the investment cycle. It changes how projects are identified, how feasibility is tested, how institutions are mapped, how capital is structured, how risks are allocated and how outcomes are measured. It also changes the meaning of infrastructure success. A completed asset is evidence that construction occurred. A reliable service that lowers costs, increases utilisation, mobilises investment and expands production is evidence that the economic system changed.

TAY's analytical proposition is therefore deliberately narrower than the claim that infrastructure is always transformative. Infrastructure becomes transformative when the surrounding system can repeatedly convert physical capacity into productive capability.

The unresolved question is not whether infrastructure matters. It is which combinations of assets, institutional reforms and financing structures produce the greatest additional economic value in a specific location and sector. That requires country-level and sector-level evidence, disciplined project preparation and continuous measurement after construction.

The practical implication is clear: the quality of an infrastructure pipeline should be judged not only by how much capital it can absorb, but by how convincingly each investment can explain the chain from constraint → service → cost and risk → investment → production and trade → measurable economic outcome.

External evidence and TAY interpretation are deliberately separated. The World Bank evidence informs the relationship between infrastructure stocks, investment efficiency, costs and growth outcomes. AfDB evidence informs the relationship between infrastructure, industrialisation and productive capabilities in Africa. PPIAF evidence informs the importance of project preparation, pipeline development, bankability, risk allocation and institutional capability in infrastructure finance. The transmission-mechanism framework, the emphasis on productive capacity as the principal public-facing test, and the decision framework are TAY's analytical synthesis rather than quotations from those institutions.

The assessment does not imply that infrastructure investment is sufficient for industrialisation or productivity growth. Human capital, competition, technology, institutions, macroeconomic stability, access to finance and market demand can be binding constraints independently or simultaneously.

There is also a measurement and attribution risk. Infrastructure outcomes can take years to emerge, while multiple reforms and market changes can occur at the same time. Project-level indicators should therefore be interpreted alongside sector and macroeconomic evidence.

The productive-capacity framework can also become too broad if every development constraint is treated as part of an infrastructure project. TAY's response is to retain a binding-constraint discipline and identify the specific mechanism the investment is intended to change.

A further risk is financing substitution: concessional or blended capital can make weak projects appear viable without fixing their underlying economics. Capital architecture should therefore follow the economic function, not the availability of finance.

The unresolved analytical question is which combinations of assets, institutional reforms and financing structures produce the highest additional economic value in a given location. That requires country- and sector-specific evidence rather than a universal infrastructure formula.