In Nepal, imagine an industrial company considering solar power, battery storage and intelligent energy management. The engineering case may be credible. The system could reduce fuel costs, limit disruption from unreliable power and lower emissions. The investment can still fail to happen. A commercial lender sees unfamiliar technology, uncertain savings and a borrower whose main business is manufacturing, not energy. The factory sees high upfront cost and operational risk. The technology provider can design the system but cannot carry the customer’s financing problem. An investor wants a portfolio, while the first factory appears as a bespoke transaction. Everyone may agree that the project is promising. No one wants to be first. This is where grants can be unusually valuable. It is also where they are often poorly used. A grant can pay for the equipment. That makes the first project easier. It may also teach the market that clean industrial energy happens only when a donor buys it. Or a grant can pay for the work that allows many investments to happen: energy audits, engineering design, performance measurement, legal templates, safeguards, aggregation and the evidence lenders need to price risk. The consequential choice is between subsidising an asset and creating an investment pathway. The first produces a visible installation. The second may produce fewer panels in the first year, but a repeatable route for the next ten factories. I favour using grant capital to remove shared uncertainties and transaction costs, while reserving commercial or concessional capital for assets that can generate savings or revenue. This is not a rule against capital subsidy. Some technologies and markets need it.
The point is to know what the grant is buying. A useful grant should purchase at least one of four things. It can buy knowledge that others can use: verified performance, cost curves, failure data and operational lessons. It can buy coordination that no individual investor will finance: assembling factories, lenders, technology firms and public agencies around a common pipeline. It can buy capability: helping firms prepare credible projects and helping local banks assess them. It can buy early risk: absorbing a defined uncertainty that prevents capable investors from participating, with a plan for that support to decline. These functions differ from making an unattractive project look temporarily affordable. The difference becomes clearer at portfolio level. Suppose five to ten factories share similar problems: power reliability, heat stress, inefficient motors, high diesel use and limited access to long-term finance. Evaluating each separately is expensive. Standards differ. Data are weak. Deal sizes may be too small for international capital. A facility could use grants for common audits, technical assistance, measurement and portfolio preparation. Concessional finance or guarantees could cover early perceived risk. Local banks or development finance institutions could fund equipment. Factories could repay from savings where the economics allow it. The grant sits before the investment. Its purpose is to make the risk legible, the projects comparable and the pipeline large enough to finance. This principle reaches beyond energy. In climate-resilient agriculture, grants can finance farmer organisation, agronomy evidence and buyer coordination before working capital becomes viable.
In urban resilience, grants can pay for risk assessment and project preparation before municipal or private investment. In health, philanthropy has sometimes created demand or evidence for products that markets would otherwise neglect. The most powerful grants do not merely fill a funding gap. They change the shape of the opportunity. But blended finance language can hide weak economics. Almost any project can be described as catalytic if future investment remains conveniently unspecified. The word “de-risking” can mean transferring private risk to the public without creating public value. So every grant-before-investment proposition should answer five questions. What specific uncertainty prevents investment today? Why is a grant the right instrument for that uncertainty? Who benefits from the knowledge or capability created? What commercial or public capital could follow if the uncertainty is reduced? Under what conditions will grant support decline or stop? If these questions cannot be answered, the grant may be subsidising hope. There is also a moral choice inside the financial one. Grant money is scarce. Every dollar used to prepare an industrial investment is unavailable for direct support to vulnerable communities. The case for using it must therefore rest on additionality: without the grant, would socially valuable investment fail, shrink or arrive much later? The answer should not be assumed. Large companies may be able to finance improvements themselves. Technology providers may have an incentive to fund demonstrations. Banks may describe projects as risky when the real issue is their own limited effort or unfamiliarity. Good programme design tests these possibilities before offering subsidy.
My judgment is to use grants where the resulting capability or evidence becomes a shared asset, where the public benefit is material, and where there is a plausible route to reduced subsidy. I would change my mind in three situations. First, if factories captured nearly all the benefits and had adequate balance sheets, they should pay for preparation and equipment. Second, if several well-designed pilots failed to change lender behaviour or generate a viable pipeline, continued grant spending would be difficult to justify. The binding constraint might be regulation, management incentives or weak underlying economics. Third, if direct public investment produced greater and fairer benefits, the effort to attract private capital might add needless complexity. The aim is not to make every resilience problem investable. Some public goods should be publicly or philanthropically financed. The aim is to use each form of capital for the work it can do best. There is a final test: whether the grant creates information that changes behaviour beyond the original transaction. If performance data remain private, legal templates cannot be reused, and every lender repeats the same assessment, the demonstration has produced an asset but little infrastructure. Shared learning needs rules: what data can be disclosed, who validates it, how commercial confidentiality is protected, and who pays to maintain the standard. These details sound administrative. They determine whether the second deal becomes easier or begins again from zero. The sequencing also matters. Technical assistance delivered after financing decisions may become decoration. A guarantee offered before weak projects are prepared can protect poor underwriting. Grants, concessional capital and commercial finance should enter when their particular constraint appears. Blending them in one facility does not mean using them at the same time or for the same risk. A grant is most catalytic when the next project needs less of it.
Leave a comment