Purpose - The purpose of this paper is to shed new light on the debate about the appropriateness of the Kaizuka rule, a Samuelson type of efficiency rule for public inputs, for the provision of firm-augmenting public inputs. Firm-augmenting public inputs are commonly included in public infrastructure modelling. Design/methodology/approach - In the microeconomic social surplus framework, and assuming perfect competition, the paper analyses how firm-augmenting public inputs should be provided in order to maximise the welfare of consumers and producers. For this purpose, the paper develops a social surplus efficiency rule, i.e. the Boadway rule. Afterwards the question what the characteristics of firm-augmenting public inputs mean for its efficient provision is examined. Findings - The findings show that under perfect competition an omniscient government is unable to efficiently provide firm-augmenting public inputs due to the characteristics of firm-augmenting public inputs but not due to inappropriate efficiency rules. Research limitations/implications - The findings show that future research would be ill advised to model public infrastructure as a firm-augmenting public input. Practical implications - Policy conclusions drawn from models that include firm-augmenting public inputs, such as fiscal competition and endogenous growth models, should be reconsidered. Originality/value - The paper makes a strong case that firm-augmenting public input is not a viable concept for modelling public infrastructure. Rather, firm-augmenting public inputs are similar to free goods.