Heterogeneous Innovation Persistence: Evidence From Uruguayan Firms

Machado, Maximiliano

Resumen:

This research addresses the persistence in innovation results for Uruguayan firms in the period 2004 –2015. Using panel data from the Survey of Innovation Activities, persistence in products and process innovations is estimated, investigating also heterogeneous effects in size and sectors. The estimations were defined according to the methodology proposed by Wooldridge (2005) to control for firms’ individual heterogeneity. The findings indicate that innovation results are not persistent in Uruguayan firms, showing null and negative effects of previous innovation on future innovation, indicating that the probability of innovating in t is non-affected or reduced for firms that innovated in t-1. Delving into these results, which is not usual in the literature in the field, the effects of the t-2 lag are estimated. Results indicate that innovating in t-2 increases the likelihood of persistence in innovation in t. This fact suggests that the Uruguayan firms innovate intermittently, contrary to what the literature states, arguably following an uneven innovation trajectory. Such results distance from empirical evidence for developed countries; although, they are in line with results for countries in the region and the case of Portugal. The effects may be related to the high costs of innovating continuously and the scarce relation with the environment, factors in which Uruguayan firms are lagging in relation to firms in developed countries.


En este documento se investiga acerca de la persistencia en resultados innovadores para firmas uruguayas en el periodo 2004 –2015. Empleando datos de panel de la Encuesta de Actividades de Innovación, se estima el grado de persistencia en productos y procesos indagando sobre efectos heterogéneos por tamaño y sector. Las estimaciones se hacen siguiendo la metodología de Wooldridge (2005) para controlar por heterogeneidad individual de las firmas. La evidencia obtenida muestra que los resultados innovadores no son persistentes, encontrando efectos nulos y negativos de innovaciones pasadas sobre futuras, indicando que la probabilidad de innovar en t no se ve afectada o se reduce para firmas que innovaron en t-1. Este resultado es contrario a lo habitual en la literatura especializada y, para profundizar su análisis se estudia el efecto de un segundo rezago en t-2. Los resultados indican que innovar en t-2 incrementa la probabilidad de innovar en t. Esto sugiere que las firmas uruguayas tienen un comportamiento innovador intermitente que puede redundar en una trayectoria de innovación errática. Estos resultados se distancian de la evidencia empírica usual para países desarrollados, aunque se alinean con algunos resultados para países de la región y con el caso de Portugal. Los altos costos de mantener una conducta innovativa constante y el escaso relacionamiento con el medio aparecen como algunas de las causantes de esta ausencia de persistencia innovadora.


Detalles Bibliográficos
2021
Innovación
Persistencia
Datos de panel
Uruguay
Innovation
Persistence
Panel data
ECONOMETRIA
ECONOMIA INDUSTRIAL
CAMBIO TECNOLOGICO
Inglés
Universidad de la República
COLIBRI
https://hdl.handle.net/20.500.12008/27956
Acceso abierto
Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0)
Resumen:
Sumario:This research addresses the persistence in innovation results for Uruguayan firms in the period 2004 –2015. Using panel data from the Survey of Innovation Activities, persistence in products and process innovations is estimated, investigating also heterogeneous effects in size and sectors. The estimations were defined according to the methodology proposed by Wooldridge (2005) to control for firms’ individual heterogeneity. The findings indicate that innovation results are not persistent in Uruguayan firms, showing null and negative effects of previous innovation on future innovation, indicating that the probability of innovating in t is non-affected or reduced for firms that innovated in t-1. Delving into these results, which is not usual in the literature in the field, the effects of the t-2 lag are estimated. Results indicate that innovating in t-2 increases the likelihood of persistence in innovation in t. This fact suggests that the Uruguayan firms innovate intermittently, contrary to what the literature states, arguably following an uneven innovation trajectory. Such results distance from empirical evidence for developed countries; although, they are in line with results for countries in the region and the case of Portugal. The effects may be related to the high costs of innovating continuously and the scarce relation with the environment, factors in which Uruguayan firms are lagging in relation to firms in developed countries.