Contagious Social Practice?

This paper explores the usefulness of spatial innovation diffusion theory for understanding the observed evolution of the social practice of attaining public support as a substitute for working. The empirical example chosen is sick leave. Several European countries have experienced a striking increase in costs for sick leave during the recent years, nowhere however as dramatic as in Sweden. Public costs for sick leave have doubled since 1997 while general health and life expectancy continually improve. No more than a small percentage of this change can be assigned to compositional changes in the population and labour market with respect to age, sex, education, location and sector of employment. Neither has economic benefit conditions changed substantially during this period. What essentially remain, as possible explanations, are changes in attitudes and practices among beneficiaries and physicians. One can observe substantial regional differences in the level of sick leave and apparent traces of spatial autocorrelation in the pattern. The spatial diffusion hypothesis was examined by help of a multivariate analysis based on socio‐economic panel data including sick leave benefits and location for each individual in the labour force 1996–2000. The analysis indicates that the neighbourhood effect is substantial on the local (290 municipalities) and the regional (21 counties) level. Largely, the regional diffusion effect remains when also controlling for more persistent average differences between counties. A semi‐public spatial attitude diffusion process, mainly operating at municipality and county levels, seems to be responsible for a substantial part of the observed increase in cost for sick leave in the period 1996–2000.     

The analysis of innovation processes

Methods are described for modelling both the speeds and the driving forces of innovation processes, namely Critical Path Schedules and Cash Flow Curves respectively. The differences between innovations in modern industrial conditions and simpler societies are discussed, and the models applied to early innovations, but the data are too fragmentary to make quantitative models, although they can be used qualitatively to show general trends. In early societies the speeds were very slow but benefit/cost ratios very high, whereas the reverse tends to be true on average for modern innovations. Long delay times are often found in early innovation processes and these are determined by a wide variety of cultural factors. Seldom has the progress of the technology itself been on the “Critical Path” of innovation processes. The “Critical Path” and “Cash Flow” methods can also be used to study other, non-innovative, complex processes.