Modelling household responses to energy efficiency interventions

1. Modelling household responses to energy efficiency interventions Ian Durbach (ian.durbach@uct.ac.za) Stephen Davis (sjdavismail@gmail.com)

2. Aims • To include consumer preferences in a household electricity consumption model • To integrate real-world data into the system dynamics modelling process

3. use use

4. Hypothesis 1: For price interventions, smaller rebound effects are experienced in households for whom cost is relatively important. Hypothesis 2: Increasing the importance of the impact objective at the expense of the cost objective would reduce the impact made by price-based interventions. Hypothesis 3: For efficiency interventions, greater rebound effects are experienced in households for whom cost is relatively important. Hypothesis 4:Smaller rebound effects are experienced if an efficiency intervention involves a technology that delivers a satisfying service. Hypothesis 5: Rebound effects occurring as a result of unsatisfactory new technology are smaller if more weight is placed on environmental issues.

5. Zanemvula solar water heater project 09/2008 03/2009 06/2009 08/2009 People start moving into upgraded houses End of study period Solar water heaters installed (78/417 households) Survey 1 (n=417) Survey 2 (n=78)

6. Hypothesis 1: For price interventions, smaller rebound effects are experienced in households for whom cost is relatively important. Hypothesis 2: Increasing the importance of the impact objective at the expense of the cost objective would reduce the impact made by price-based interventions. Hypothesis 3: For efficiency interventions, greater rebound effects are experienced in households for whom cost is relatively important. Hypothesis 4:Smaller rebound effects are experienced if an efficiency intervention involves a technology that delivers a satisfying service. Hypothesis 5: Rebound effects occurring as a result of unsatisfactory new technology are smaller if more weight is placed on environmental issues. <0 -1.01 (p=0.009) >0 +0.53 (p=0.003) >0 +0.52 (p=0.01) <0 -0.02 (p=0.11) <0 -1.48 (p=0.007)

7. Aims • To include preferences in a household electricity consumption model • Quite complex dynamics can be explained by relatively simple model of consumer “preferences” • To integrate real-world data into the system dynamics modelling process • Confidence in model can be established by validation against real-world data, even in a fairly “qualitative” mode