On profit differentials between persistent and occasional innovators: new evidences from a random coefficient treatment model

The paper studies the medium term effect of being a persistent (occasional) innovator on firm economic return within a "counterfactual" setting using a random coefficient model. This approach allows us to assess not only the point effect of a persistent /occasional innovation strategy on profitability, as in standard regression settings, but the entire distribution of it. We exploit a nine years (1998-2006) longitudinal dataset of Italian manufacturing firms obtained by a merging of the slat three waves of the Capitalia/Unicredit surveys. Results show a strong better economic performance of the group of firms that continuously implement their innovating capacity and output. Also occasional innovation produces good operating profit margin (OPM) differentials, although we estimate a difference with the persistent behavior of about three percentage points lower. Differences between persistent and occasional innovators are also enlightened at a dynamic level: we found that persist innovation allows for a dynamic advantage against occasional "first time only" innovation strategy. Moreover, the analysis of the idiosyncratic distribution of the effect, based on the random coefficient model allows us to inspect what factor lead to be persistent innovators and we identify the "best performers" among them. These champions are characterised by a large stock of accumulated knowledge, a large size and operate in more concentrated markets. This result confirms what we have found in the literature on innovation persistence: dynamic capability building can be found mainly when a mechanism of increasing return to scale is operating and this is mainly present in few leading companies.