Innovation and financial constraints centralised and decentralised economies

Because of the "too-early-to know" feature of innovation, the promotion of innovation in a given system critically depends on the screening mechanisms which terminate inefficient projects. Information imperfection together with sunk costs may cause commitment problems which relax financial constraints and thus disable the screening capability in a centralized economy. This leads to the "too-late-to stop" consequence, that is, the continuation of inferior projects. As an optimal organizational response to this problem, centralized economies often use bureaucratic procedures based on prior information to pre-screen projects. However, pre-screening mitigates the problem at the cost of rejecting promising projects and delaying innovation. In the presence of externalities when returns are non-addictive, the market screening effect may compensate for the inefficient duplication effect, thus innovation in decentralized economies may still be more efficient than in centralised economies. Because of the commitment problem, the optimal number of projects in a centralized economy may be less than that in a decentralized economy. Furthermore, with smaller projects that have a higher probability of failure, a centralized economy starts even fewer parallel research projects than does a decentralized economy, which thus reduces the chances of success. The failure of the computer industry in centralized economies illustrates this point.