Companies want to understand attitudes, interests, and rate of new product adoption. Economists from Veritas have combined survey research, econometric modeling, and dynamic simulation techniques to evaluate new product adoption for pharmaceuticals, medical devices, electric vehicles and residential solar.
Electric Vehicle Adoption Survey
The Electric Power Research Institute (EPRI) and project funders were interested in understanding electricity customers’ perceptions and expectations of electric vehicles in order to evaluate the role of utilities in supporting the adoption of electric vehicles. Veritas conducted a survey developed to understand these issues. The main component of the survey asked respondents to score how likely they were to purchase four types of vehicles: gas, hybrid electric, plug-in hybrid electric, and battery-only electric. Respondents were asked to indicate their purchase likelihood of these vehicles by allocating 100 points across one, two, three, or all four vehicles.
Results indicated that EV purchase likelihood is largely influenced by the following factors: previous knowledge of EVs through research in the past 6 months, demographics, and driving behaviors. The survey also presented respondents with the option of 1 of 3 charging plans and willingness to pay for a number of charging options. Responses on expected conditions, purchase likelihood, and associated demographics were used to inform development of the initial adoption stages of a structural model of electric vehicle adoption.
Electric Vehicle Adoption Simulation Model
EPRI builds and maintains the US Regional Economy, Greenhouse Gas, and Energy (US-REGEN) model. This model contains a Transportation Module which includes projections of US car purchasing and driving behaviors. EPRI was interested in embedding a structural (e.g. responsive to fuel prices) dynamic model of electric vehicle adoption into this Transportation Module. Veritas created a module which combines results from the EPRI/Veritas electric vehicle preference survey with a revealed preference model of vehicle choice to develop customer preferences for new vehicle purchase including electric vehicles.
Choice model results are incorporated into a stock turnover model following Struben and Sterman (2008) innovators and imitators structure. The choice modeling component of the adoption model identifies new car purchases using a mathematical simulation of vehicle choices given consumer characteristics, preferences for vehicle attributes, and vehicle choice sets. Preferences for non-Electric Vehicle characteristics are based on a statistically estimated model using vehicle sales and customer attribute data. Vehicle preferences are tied to the unique characteristics of electric vehicles that through parameters that can be calibrated based on reduced-form Electric Vehicle estimates and relationships identified in the electric vehicle survey research Veritas conducted for EPRI.
Pharmaceutical and Medical Device Adoption
Veritas staff were involved in some of the early efforts using survey research, preference based econometric modeling, and simulation to characterize adoption of pharmaceutical and medical devices. Efforts include developing a decision model for evaluating the effect of information on toxicity, efficacy, cost, and alternative treatments on anticipated adoption rates of developmental compounds (Bingham, Johnson, and Bell 1998), linking health state experiences to adoption estimates from preference based pharmacoeconomic studies (Bingham and Smith 2001) and developing a dynamic outcomes based choice model of new pharmaceutical adoption that incorporates Bayesian updating in a Markov model of outcomes (Bingham, Johnson, and Miller 2001)
Residential Solar Adoption
Residential solar is becoming increasingly prevalent in the U.S. Because residential solar penetration has implications for electrical loads, the Electric Power Research Institute and supporting power companies wanted to understand adoption rates under different conditions. Veritas integrated baseline penetration information with econometrically modeled survey data to predict solar adoption rates under various conditions and in different locations.
Veritas developed and administered the survey to collect the data that underlies the econometric model. The model data arose from a discrete choice experiment of 7,000 electricity customers in seventeen service territories throughout the United States. Results are incorporated into a dynamic residential solar adoption model for each service territory that allows conducting adoption simulations under various conditions of electricity costs, solar system costs, and subsidies.