Search News EUKLEMS home Top picture

EU KLEMS Project

EU KLEMS Working Paper Series

No. 32

Bernd Görzig, Martin Gornig and Axel Werwatz (2008) The impact of introducing new products on firm productivity. Evidence from German manufacturing firms , (June 2008)


Globalization offers new challenges for manufacturers to flexibly adjust their product range to the ever changing patterns of demand and technology. Introducing new products is a primary measure to capitalize on new market opportunities and to temporarily earn quasi-monopolistic rents (Penrose 1959; Chandler 1978). Empirical research on the impact of new products on firm performance suffers from the lack of a generally accepted definition of what constitutes a “new” pro¬duct. Some researchers rely on the (self-) assessment of firms whether they have introduced a new product (Peters 2008). An alternative way to observe and study the introduction of new products at the firm level is offered by the firm census system of official statistics. Statistics Germany asks manufacturers in its “Production census“ to classify their output according to the European PRODCOM-list, which includes about 6000 products. We have matched this output information firm-by-firm to information on manufacturers‘ performance from the „Cost census“, yielding a data set of roughly 7700 manufacturing firms observed between 1995 and 2001. We use the combined information on their product range, gross value added and employment to study the impact of changing the product range on productivity and performance (Görzig/Gornig 2007). To separate the impact of product policy from those of other determinants of productivity we employ a decomposition approach suggested by Nopo (2004) as a nonparametric extension of the widely-used Oaxaca-Blinder decomposition (Blinder 1973; Oaxaca 1973). We decompose the observed “raw” productivity differences between modifiers and non-modifiers of the product range into a “structural“ and a “behavioural“ component. The former contains the part of the raw difference attributable to the different composition of modifiers and non-modifiers with regard to employment size and sector. The latter is the “pure” productivity difference between modifiers and non-modifiers of comparable size and sector. We find that almost 30% of all firms introduced at least one new product into their product range between 1995 and 2001. These modifiers enjoyed an average productivity advantage in 2001 of 13% over non-modifiers. The data also shows a much higher variance of observed productivity among modifiers, pointing to the risk involved in this strategy. While some firms do reap the intended above-average rent, numerous other firms don’t – either because of a (temporary) loss in efficiency from readjusting production (Montgomery 1985) or because the new products failed to find a market.

  print version (pdf, 288 KB)

Go to top of page

This project is funded by the European Commission, Research Directorate General as part of the 6th Framework Programme, Priority 8, "Policy Support and Anticipating Scientific and Technological Needs".

Mail to EU KLEMS

Reactions to this page
Last changed on: September 12 2008