EEP 143 Lecture 19 Diffusion and Adoption Outline Introduction Social and economic determinants S Curve Example Real life examples Economic determinants of diffusion Case Study I The Dynamo More economics option value Case Study II Hybrid Corn Case Study III Green Revolution Introduction Diffusion can be perceived as the spread of an innovative product throughout the economy Examples of innovative products tractor computers cell phones Adoption is diffusion from the users point of view decision whether to buy the innovation or not The diffusion process throughout innovations has shown fairly consistent behavior In the history of diffusion of many innovations one cannot help being struck by two characteristics of the diffusion process its apparent overall slowness on the one hand and the wide variations in the rates of acceptance of different inventions on the other Rosenberg 1976 p 191 Sociology Attributes influencing adoption 1 Relative advantage of innovation 2 Compatibility with industrial and or social norms 3 Complexity 4 Trialability especially at affordable scale 5 Observability ease of evaluation See Rogers E M The Diffusion of Innovations 1995 Sociology Attributes influencing speed of adoption 1 Nature of decision collective individual or top down 2 Communications channel mass media or personal 3 Social system norms interconnectedness 4 Promotional efforts of change agents advertisers extension agents pharma detailers See Rogers E M The Diffusion of Innovations 1995 Economic Determinants of Diffusion The faster agents adopt a specific innovation the faster is the diffusion of that innovation Since adoption of an innovation requires investment the rate of diffusion will be influenced by Benefits Remember determinants Costs of the NPV Uncertainty Institutions and Environment Economic Determinants of Diffusion Benefits in turn depend on Competitiveness of substitute technologies Tractors vs Horses Networks and network standards Experience observation trial and learning Costs depend on Price of the new technology Complementary investments If a firm wants to buy computers then she also has to try to train her personnel Economic Determinants of Diffusion Uncertainty can be found on many aspects New technology can be insufficiently understood There can be a natural risk aversion Costs of adoption are obvious they are in t 0 but benefits are to come in the future There might be uncertainty for the benefits There are many environmental characteristics that can affect the speed of diffusion Size and market power of adopters Size and market power of suppliers of the innovation Economic Determinants of Diffusion Also institutions can Accelerate diffusion by trying to promote standards and or through policy promote the technology itself For instance if you want to promote hybrid cars let the owners park for free in the parking meters this is what I think is happening or is going to happen in Greece Delay diffusion especially if there is political cost If there are powerful lobbies that the new innovative product will harm their interests maybe the government wants to keep the lobbies happy for instance GMOs in US vs the major agro chemical firms in the EU S Curve The proportion of people at time t that have adopted the innovation can be expressed by the following equation a y t b ke at Even though there are much more complicated models this equation has been very insightful in the understanding the economics of diffusion The carrying capacity of the population is a b S Curve In other words it is the proportion of people that will eventually adopt the innovation This is true because a t y b So for instance For electricity in the US we can fairly assert that a b However for the operating system of Microsoft Windows probably a b it is fair to assert that not everyone will eventually use Microsoft Windows since a lot of people use OS Linux Unix etc Also note that the above magnitude is often called steady state because after t getting larger and larger y becomes essentially constant S Curve k is the measure of slowness rate of diffusion To interpret k note that a k 0 y b To see how the diffusion rate evolves note that at y ae 0 at 2 k b ke S Curve Example k 100 a b 1 1 0 8 almost full diffusion it takes 15 years 0 6 0 4 0 2 2 4 6 8 10 12 14 S Curve Example k 10 a b 1 1 0 8 Almost full diffusion takes 10 years 0 6 0 4 0 2 2 4 6 8 10 S Curve Real Life 100 90 80 Telephone Share 70 Refrigerator 60 Washing machine 50 40 VCR 30 20 10 0 1900 Electric Service 1910 1920 1930 1940 1950 1960 1970 Year Source Dallas Federal Reserve Bank 1980 1990 2000 Case Study I The Dynamo In the 1980s even though everyone knew that the personal computer is a major invention they were not seeing high rates of diffusion and most importantly increases in productivity Solow We see the computers everywhere but in the productivity statistics David 1990 compared the diffusion of the electric dynamo with the diffusion of computers Case Study I The Dynamo Electric dynamo electric generator converts mechanical energy into electricity using magnetism Even though now it is so common it took some time like computers to spread throughout the whole economy Case Study I The Dynamo Diffusion rate of dynamo Bronwyn Hall Case Study I The Dynamo Electric motors originally in existing group drive systems used with steam water power Does this raise capital intensity of new electric technology Is David p 357 right on this Later unit drive installations allowed factory redesign single story skylights variable power control Quality effects hard to measure in productivity stats worker safety increased Lighting of stores factories homes Commuting to pleasant housing areas via electric rail Case Study I The Dynamo Unit drive system required a whole new set of skills by the factory architects construction firms production workers Increased shift work capital utilization in petroleum paper and chemical industries Used in industrial control processes Adopter feedback increased efficiency Case Study I The Dynamo Implications for PC Rise of separate software industry Role of access for outside programmers Apple v IBM Customer feedback Compatibility and market tipping Persistence of parallel processes e g in UC budget process for years Email and communications The paradoxical effects of rapid innovation Adoption usually involve high initial fixed costs then flow of benefits If benefits are high adoption tends to be more attractive What if you