A profitable factory and a valuable project are different tests

The financial case is built from cash flows the investing entity actually experiences: capital expenditure, operating costs, revenue, tax paid, working capital movement and residual value. Its outputs are familiar — net present value, internal rate of return, payback — and its test is whether the return clears the cost of capital and the risk.

Two errors are common here. The first is including costs or benefits that the entity does not bear or receive. The second is omitting cash effects that the entity does — working capital being the most frequently overlooked, particularly during a ramp period when inventory and receivables both build.

Taxes are a particular trap. Tax paid is a genuine cash cost to the investor and belongs on the financial ledger. It is also a transfer to government, which creates no net resource use for the economy as a whole, so it is treated differently on the economic ledger. Including it in both, or in neither, produces a distorted comparison.

The economic case attempts to capture effects that fall outside the investing entity: emissions, congestion, resource depletion, employment effects and, where relevant, wider productivity spillovers. These are real effects, but they are also where the analysis most easily goes wrong.

The first hazard is double counting. If a benefit is already captured in the financial cash flows — a cost saving, for example — it should not be added again as an economic benefit. The economic appraisal is measuring the total effect on the economy, of which the private effect is usually a part, not an additional item.

The second is the treatment of transfers. A subsidy is a transfer from the public purse to the project. It is a genuine cash inflow to the investor and therefore belongs on the financial ledger, but it is not a net gain to the economy, because the resources came from elsewhere. Counting it as an economic benefit alongside the activity it funded counts the same money twice.

The third is the valuation of non-market effects. Shadow prices, willingness-to-pay estimates and similar techniques are legitimate, but they are estimates with ranges, and they should be presented as such rather than as measured quantities.

The distinction is not academic. It determines who the analysis is for and what follows from it. A financially unviable project with strong wider benefits is a candidate for public support, blended finance or a different contractual structure — not for abandonment. A financially attractive project with significant unaccounted external costs may need mitigation or a charge, not approval.

It also clarifies what an honest appraisal can and cannot claim. A commercial appraisal cannot establish that a project is socially worthwhile, and an economic appraisal cannot establish that it will attract capital. Presenting one as though it were the other is the most common way an otherwise competent analysis becomes misleading.

For a manufacturer considering a significant investment, the practical implication is to run the private case rigorously first, because that determines whether the project can proceed at all. The wider case is then useful for conversations with public bodies, for understanding exposure to future regulation, and for identifying where support might legitimately change the decision.