The paper introduces a reform trajectory we call revolutionary incrementalism in which partial and incremental measures add up to profound transformation. Recent advances in economic theory demonstrate that growth is not hard to start: it almost starts itself, somewhere, sometimes. But keeping it going is not easy: doing so requires attention to the context of growth-binding constraints and situation-specific ways to resolve them. The same goes for institutions: it is almost always possible to find some that are working. The issue is using the ones that work to improve those that don't. The thrust of the proposal is to rely on variation within existing institutions as the Archimedean lever with which to leverage reform and change. India's public sector record for implementing and coordinating innovation efforts can be notoriously fragmented and inefficient but there are some parts that perform better than others, and there are recognized pockets of excellence virtually within every ministry or public sector organization. The same internal diversity is even more visible in the private sector.
innovation systems heterogeneity of institutions radical incrementalism search networks open economy industrial policy