Socrates Kypreos

Endogenizing R&D and Market Experience in the "Bottom-Up" Energy-Systems ERIS Model

ERIS, an energy-systems optimization model that endogenizes learning curves, is modified in order to incorporate the effects of R&D investments, an important contributing factor to the technological progress of a given technology. For such purpose a modified version of the standard learning curve formulation is applied, where the investment costs of the technologies depend both on cumulative capacity and the so-called knowledge stock. The knowledge stock is a function of R&D expenditures that takes into account depreciation and lags in the knowledge accumulated through R&D. An endogenous specification of the R&D expenditures per technology allows the model to perform an optimal allocation of R&D funds among competing technologies. The formulation is described, illustrative results presented, some insights are derived, and further research needs are identified.

Emissions trading and technology deployment in an energy-systems “bottom-up” model with technology learning

Abstract An important criterion in the analysis of climate policy instruments is their ability to stimulate the technological change necessary to enable the long-term shift towards a low-carbon global energy system. In this paper, some effects of emissions trading on technology deployment when technology learning is endogenized are examined with a multi-regional “bottom-up” energy-systems optimization MARKAL model of the global energy system. In this framework, due to the action of spillovers of learning, imposing emission constraints on a given region may affect the technology choice and emissions profiles of other (unconstrained) regions. The effects depend on the geographical scale of the learning process but also on the presence of emissions trading, the regions that join the trade system and their timing for doing so. Incorporating endogenous technology learning and allowing for spillovers across regions appears as an important mechanism for capturing the possibility of induced technological change due to environmental constraints in “bottom-up” models.

Linking energy system and macroeconomic growth models

We compare two alternative approaches for coupling macroeconomic growth models (MGM) and energy system models (ESM). The hard-link approach integrates the techno-economics of the ESM completely into the MGM and solves one highly complex optimisation problem. The soft-link leaves the two models separate and energy supply functions are integrated into the MGM that are derived from the optimal solution of the ESM. The energy supply functions relate the price of energy computed with the ESM to the quantity of energy computed with the MGM. An iterative process exchanges price-quantity information between the models. Hence, the soft-link leads to an energy market equilibrium. But energy supply functions do not consider variable interest rates that influence the energy supply functions. This is due to the fact that ESMs are partial models that assume an exogenous interest rate; however the interest rate is computed endogenously in MGMs. This missing interaction leads to a capital market dis-equilibrium in the soft-link compared to the hard-link approach inducing a mis-allocation of investments. Extending the soft-link approach by also considering the time variable interest rate of the MGM does not improve the results. Though the computational complexity is greater the hard-link approach assures simultaneous energy and capital market equilibrium.

ERIS: A model prototype with endogenous technological change

This paper describes the ERIS Model Prototype, developed within the EU-TEEM project as a flexible tool to study different modelling approaches on the endogenisation of technological change in energy optimisation models. The basic formulation and several variants are presented. Some illustrative results and insights obtained with the prototype are discussed and some perspectives for its future development outlined.

A MERGE Model with Endogenous Technological Progress

We have developed a new version of the MERGE model, called MERGE-ETL, to analyse the dynamics of endogenous technological learning (ETL) in the energy system. This paper describes the basic formulation of MERGE-ETL, the solving techniques used for this model and some first numerical results in the context of policies designed to mitigate global climate changes.

Multi-regional technological learning in the energysystems MARKAL model

This paper describes the implementation of multi-regional endogenous technological learning in the energy optimisation MARKAL model. A mapping procedure is implemented to group learning technologies inside one region or across several regions in a flexible way, in order to allow them to learn together. The approach is described and an illustrative example examining the response of a multi-regional global electricity generation system is presented. The multi-regional learning framework allows the examination of the spatial interactions and mechanisms that affect the technological learning processes in global energy systems. The mutual interactions between the learning and emission trading mechanisms are highlighted. Although emphasis is given to energy modelling, some policy insights can be gained. The results highlight the importance of fostering international cooperation to stimulate the learning process of emerging energy technologies.

Flexible Carbon Mitigation Policies: Analysis with a Global Multi-Regional MARKAL Model

The Swiss National Centre of Competence in Research (NCCR-Climate) explores the predictability, variability and risks of climate change and the socio-economic response to it. The Paul Scherrer Institut (PSI) and the University of Geneva contribute to this programme by using models to simulate the impacts of policies for climate change mitigation. This study quantifies the benefits of several policies enhancing the flexibility of carbon dioxide (CO2) mitigation, with emphasis placed on emissions trading, optimal timing paths and support for learning-by-doing (LBD) in the use of low-carbon technologies. We present illustrative results for a “Soft-landing” scenario, which imposes a CO2-emission stabilization target that is consistent with stabilizing CO2concentration at 550 ppmv in the long run. This analysis has been conducted with the Global MARKAL Model (GMM), which is a multi-regional, “bottom-up”, partial equilibrium energy-system model with endogenized technology learning (ETL). Incorporation of flexible CO2mitigation policies leads to significant reductions in energy-system costs and marginal costs of CO2abatement as well as increasing diffusion of advanced low-carbon technologies. In the future, an extended GMM model could be linked to a climate model (e.g., C-Goldstein, Marsh et al., 2002) to implement an Integrated Assessment Model (IAM) that would allow examining impacts of climate change.

A post-Kyoto analysis with the ERIS model prototype

Electricity generation technologies are examined in a global context with a multi-regional version of the ERIS model prototype with endogenous technological learning curves, developed within the EU/TEEM project. Impacts of Kyoto-like CO2 constraints are analysed considering the effects of allowing or not trade of emission permits. Complementary stochastic analyses addressing the uncertainty of emission constraints, demand and learning rates and a preliminary assessment of the effects of the geographical scale of learning are also presented. When technology dynamics are endogenous, mitigation policies stimulate technological learning of emerging marginal low carbon technologies driving the model to their early deployment. Trade of emission permits allows some of the constrained regions to take more moderate actions, but provides opportunities for penetration of learning technologies in different regions, contributing to their long term cost competitiveness. Early action appears to be effective in terms of long term costs and emission profiles. Uncertainties in emission targets and demands may stimulate technological learning as a preparation for future contingencies.

Incorporating different endogenous learning formulations in MERGE

This paper presents the implementation of endogenous technological change, following either a one-factor or a two-factor learning curve, in the MERGE model. We have indeed developed a new version of this model, called MERGE-ETL, to analyse the dynamics of endogenous technological learning (ETL) in the energy system. This paper describes first the basic formulation of MERGE-ETL, then the solving techniques used for this model, and presents finally some numerical results in the context of policies designed to mitigate a global climate change.

Swiss energy taxation options to curb CO2 emissions

This study offers insights into the design of economically efficient policies to curb carbon dioxide (CO2) emissions in Switzerland and in other European countries. The method uses a model of the energy system to investigate various options for taxation to reduce CO2emissions. This study proposes as a first option the introduction of a ‘hedging tax’, that balances the risks of delaying measures to reduce CO2 emissions against those of premature reduction measures. It then assesses multinational policy options and considers as a second alternative international co-operation to curb joint CO2 emissions by means of a uniform tax applied in different countries. The simulation of such a strategy among three European countries (Switzerland, the Netherlands and Belgium) suggests that there may be significant benefits to be gained when CO2 reduction takes place in the countries where it is relatively cheap to do so.