Changes in the Speed of New Product Diffusion: Mere Acceleration versus Economic and Demographic Conditions

It is a popular contention that products launched today diffuse faster than products launched in the past. However, the evidence of diffusion acceleration is both scant and mixed. Moreover, the methodology used in previous studies has a number of disadvantages. This study re-assesses the acceleration thesis in a single country (U.S.) for 37 consumer durables over a period of 74 years (1923-1996). To avoid some of the methodological limitations of extant research, this study takes a different approach in both measuring diffusion speed and analyzing the data. Diffusion speed is measured as the empirical hazard of diffusion. Historical change in diffusion speed is described using a variant of age-period-cohort analysis. Whether diffusion speed has increased over time and how it is causally related to some explanatory variables is tested by means of panel data regression analysis. Two main findings emerge. First, there has been only a very small systematic increase in diffusion speed over time. Second, changes in diffusion speed are affected more by changes in disposable income, unemployment, and household formation rate than by a general acceleration trend. Implications are discussed for diffusion modeling at the level of the individual product, as well as for the broader issue on how economic and demographic factors, rather than cultural change, have been the key drivers in the rise of consumption intensity. (Diffusion, New product research)