Behavioural response to investment risks in energy efficiency

Transcription

1 Meeting CA EED Sofia, 20 October 2017 Heterogeneity in consumer decision-making for energy efficiency - can target-group specific policies help to boost energy efficiency in private households? BRISKEE Behavioural response to investment risks in energy efficiency Funded in the H2020 programme

2 The discount rate dispute PRIMES 2013/EC This boils down to the question in how far discount rates used to evaluate FU- TURE policies shall reflect PRESENT individual decision making processes with rather imperfect mechanisms to include risk assessment into the discount rates. PRIMES integrates (perceived or existing) risks into the discount rates to a large degree, our scenario approach essentially uses usual capital costs, considering that there are instruments to mitigate the risks and the risk perception Fraunhofer ISI

3 Motivation - The start of the discount rate debate (Comparison of discount rates applied by different countries) Source: Steinbach et al. 2015

4 Policy Consequences of Discount Rates EU COM: Communication on Energy Efficiency July 2014

5 Example how assumptions on discount rate impact on how an energy efficiency measure is rated Left: Net present value per m 2 floor area for an investment in thermal renovation and discount rates from 0% to 15%. Right: Net present costs of direct electric heating versus heat pumps for discount rates of 3% and 9%. Source: BRISKEE

6 Geographical focus: EU-28 Survey in 8 countries 75% of EU energy consumption France Italy Germany Poland 76% of EU population Romania Spain Sweden UK Source: BRISKEE

7 Technological scope: Different purchase price categories Low cost lighting Medium cost White goods High cost Thermal retrofit and heating Source: BRISKEE

8 Micro level Energy efficiency technology adoption in households Multi country household survey on effects of time & risk preferences on EET adoption Micro econometric analysis Meso level Energy efficiency policy and household investment behavior Integration of empirical findings into models for residential buildings (Invert/EE Lab), appliances (FORECAST) Macro level Macro economic effects of energy policy Translation of results from energy modelling into input to macroeconomic modelling (ASTRA)

9 Factors underlying the IDR Preferences Exemplary factor-specific policy interventions Time preferences Rebates, loans Risk preferences Loans with fixed interest Implicit Discount Rate Reference-dependent preferences (e.g. loss aversion) Pro-environmental preferences Predictable (ir)rational behavior Bounded rationality Rational inattention Behavioral biases Policies taking into account preferences and predictable (ir)rational behavior Reactive policy intervention Social comparisons, framing, default setting Eco-labelling Minimum energy performance standards Energy taxes Feedback, default setting External barriers to energy efficiency Split incentives Lack of information / transaction costs Lack of capital Technological risk Policies directly addressing external barriers Directed policy intervention Labelling, performance certificates Energy labelling, information campaigns, energy audits Rebates, tax incentives, soft loans and low-interest Warranties credit Financial risk Rebates, tax incentives

10 CHEETAH Split incentives Transaction costs Lack of information Lack of capital Technological risk Financial risk Directed policy interventions directly addressing external barriers Implicit Discount Rate BRISKEE Time preferences Risk preferecnes Proenvironmental preferences Reactive policy interventions accounting for heterogeneity in Time preferences, risk preferences,..

11 Results Micro Analysis energy efficiency technology adoption decisions - implications for policy (1)

12 Results Micro Analysis energy efficiency technology adoption decisions - implications for policy (2)

13 Implementation of Survey findings into the Bottom-up Models INVERT (Buildings) and FORECAST Appliances

14 Scenarios for appliances in FORECAST- Residential Scenario name Current-policy scenario Intensified-measures scenario New actor-related measures scenario Explanation All Ecodesign and Labelling measures that are adopted are explicitly modeled for refrigerators, washing machines, freezers, dryers, dishwashers, stoves and lighting and are modelled as an average over technologies for televisions, set-top boxes, laptops, desktop computers, computer screens, modems/wifirouters and air conditioning. Includes all measures implemented in the current-policy scenario and assumes that minimum standards are intensified and the label is rescaled. The rescaling of the energy label is assumed to increase its effectiveness affecting both consumers and suppliers so that more efficient appliances become available earlier. New instruments affecting actors are implemented in the model (and existing actorrelevant policies increased) taking into account findings from survey.

15 Scenarios for heating and cooling in - INVERT/EE-Lab Scenario name Current-policy scenario Intensified-measures scenario New actor-related measures scenario Explanation The Current-policy scenario considers targets and measures concerning RES-H/C and energy efficiency which have been decided or already implemented. On the European level, the relevant policy implications are particularly set by the Renewable Energy Directive, the Energy Efficiency Directive, the Directive on Energy Performance of buildings, and the Ecodesign Directive. The intensified measures scenario assumes that the policies which are implemented currently are intensified; however, the policy approaches remain the same. For example, a country that currently relies on minimum efficiency standards would continue to use this approach; however, the standards would be defined in a more ambitious way. Monetary subsidies for thermal renovation and renewable heating systems are increased in all member states The new actor-related measures scenario assumes that energy efficiency policy is complemented by new policy measures affecting the discount rate of low income agents. The discount rates of low income agents are reduced to the level of median income building occupants reduction of discount rates between 2% and 5% depending on initial assumptions in member states

16 Comparison of actor-specific modelling and assumptions in the residential sector (1) Comparison of final energy for the residential sector (sum of buildings and appliances) in the BRISKEE scenarios

17 Comparison of actor-specific modelling and assumptions in the residential sector (2) Share of energy carriers in final energy demand for heating and cooling for the current, intensified and agent specific policy scenario in EU28

18 Level 3 (Macro-Analysis) 1. Generation of sectoral impulses from Invert and FORECAST 1. Investment and O&M impulse (incl. subsidies) 2. Energy expenditure impulse 2. Modelling of net effects with ASTRA 1. GDP 2. Employment 3. Trade 3. Analysis and interpretation

19 ASTRA Model

20 Impulses from energy models Impulses Investment in energy efficient heating technologies and insulation Investment in energy efficient appliances Reduced energy expenditures due to heating technologies and insulation Reduced energy expenditures due to efficient appliances Subsidies Model variable/sector affected Consumption (multiple sectors) Consumption (electronics and computing equipment) Consumption (energy sector and all other sectors) Consumption (energy sector and all other sectors) Consumption, government expenditures Consumption vector: Energy Investment Expenditure... Machinery Electronics... Constructio n... Energy...

21 Relative GDP and FTE employment development in D2 and D3 scenarios with respect to the current-policy scenario

22 The ongoing CHEETAH project: a concrete look at the design of policies that impact on decision making Micro level: Development of questionnaire using stated preferences discrete choice experiments > Household survey in 8 EU member states (online, ~2000 participants per country, representative samples) Micro econometric analysis Meso level: Integration of the empirical findings of WP 4 into the models for residential buildings (Invert/EE Lab), appliances (FORECAST) and agent based models > analyze the effects of energy efficiency policies and household energy efficiency investments on residential energy demand for all EU member states until Macro level: Translation of results from energy modelling (WP 5) into input to macroeconomic modelling > Modelling effects on GDP, employment and imports/exports until 2030

23 Where to read more... Where to hear more...briskee final conference 29 November 2017 (Brussels)

24 Many thanks for listening! Wolfgang EICHHAMMER Head Competence Centre Energy Policy and Energy Markets Fraunhofer Institute for Systems and Innovation Research ISI Breslauer Strasse Karlsruhe Germany and Utrecht University, Copernicus Institute Heidelberglaan 2, 3584 CS Utrecht, The Netherlands Phone Fax mailto:

25 Back-up

26 Motivation - The start of the discount rate debate (Discount rates applied in the PRIMES model) Source: Steinbach et al. 2015

27 Motivation - The start of the discount rate debate (Discount rates applied in the in the evaluation of the energy efficiency policy framework in the EU- Fraunhofer et al. 2014) Source: Steinbach et al. 2015