Innovating for Others: Capability, Policy, and the Ownership of Invention in Small Open Economies

Countries that invent more than they own are usually told their firms fail to capture domestic invention. This paper shows that the statistic behind that diagnosis cannot support it. In a model where patents are assigned to employers by contract, so ownership is settled in the labour market before a patent exists, the domestically-owned share of domestically-invented patents separates exactly, in a benchmark case, into a term containing only policy and institutional parameters and a term containing only domestic firm capability. Two results follow. An unconditional research subsidy is exactly neutral on ownership, because it lowers both bidders' costs identically. Attaching a recapture condition to support already given makes the same expenditure act on ownership, without new money. When capability accumulates with captured output, the economy can remain in a low steady state specialised in invention, and instruments acting on who captures a given base cannot escape it, because their effect is proportional to existing capability. Sequencing, not composition, matters. Descriptive evidence on Canada and fifteen OECD peers illustrates; identification is left to future work.