Three essays on past determinants and potential future adoption of low-carbon technology

The use of renewable fuels in producing electricity is considered as a fundamental step to decarbonise the electricity sector which provides an effective strategy to combat the climate change. Thus, this thesis examines the determinants of the non-hydro renewable technology (NHRE), investigates the potential future adoption of the NHRE and quantifies the amount of hydrogen storage which has been established as an appropriate technology to store wind electricity, required to address the excess wastes from high wind deployment. Specifically, the empirical issues addressed are: (i) to extent the configuration of the existing conventional electricity sources influences the NHRE deployment, (ii) the role of environmental policy (such as Kyoto Protocol and Feed-in-Tariffs) and natural gas profitability to ensure better comparisons with related previous research in the deployment, and (iii) the likely potential estimate bias from assuming constant effects of the deployment determinants across the OECD countries, and from not adequately addressing the endogeneity and cross-sectional dependence. The simulation chapters examine the following issues: (i) the extent to which price dynamics influence the low-carbon investment decisions in the UK electricity system, while considering the market and policy uncertainties (ii) the amount of hydrogen storage required at utility-scale level to mitigate the excess wind electricity waste, and (iii) the investment returns of deploying the hydrogen storage technology towards the UK net zero electricity transition.

The thesis commences with Chapter 2 that provides an empirical analysis of the factors underlying the observed variations in the deployment of the non-hydro renewable electricity (NHRE) generation across 35 OECD countries between 1980 and 2014. The chapter first applies the fixed-effect estimation technique to test the strong assumption of a constant elasticity of GDP, population and CO₂ emissions made in the static fixed effect literature, while allowing for additional control variables not previously considered. Then, it applies the system GMM to the dynamic panel model settings which are required, after subjecting its datasets to the panel unit root and cointegration tests. The findings from the study’s panel estimation techniques reinforce the homogenous elasticity assumption in the existing cross-country studies of the non-hydro renewable deployment as the interaction terms are not statistically significant. While addressing the endogeneity and the cross-sectional dependence in the dynamic system GMM models, the study finds that only GDP is the main driver of the non-hydro renewable electricity deployment. In addition, the significance influence associated with population and CO₂ emissions in the literature, turns to be insignificant when including the configuration of the existing conventional electricity system, the climate policy variables, and the share of natural gas rents in GDP as additional control variables. Furthermore, the system GMM results reveal that the nuclear electricity share and the potential gas electricity generation significantly influence the NHRE deployment in an opposite manner, while Feed-in tariffs (which is only considered as the related literature did for a better comparison) stimulates more uptake of the non-hydro renewable electricity. No direct effect is from signing the Kyoto treaty and the onset of its binding which might be due to bottlenecks and the capacity constraints or policy inertia but more deployment of the NHRE is observed in the later period of its binding. The research finds less deployment of the new renewable energy sources as the share of natural gas rents in GDP increases. Considering these findings, the chapter suggests that any policy that improves the economic conditions, would induce the deployment of the non-hydro renewable electricity, even with less attention on the population and the CO₂ emissions.

The first simulation chapter, Chapter 3, develops a novel agent-based model at the power plant level using the UK real electricity grid information, to incorporate a profitability value as an additional attribute for each power plant. Then, the study examines the impact of these power plants’ attributes on the low-carbon investment decision under the marginal and dual pricing methods in the context of the UK electricity system. The key findings from the developed agent-based model reveal that the role of nuclear technology towards the UK electricity decarbonisation drops to 5% at 2050, but the wind electricity share attains almost 50%, with an increase in the share of gas technology under the dual pricing system. The installed capacities of the low-carbon technologies especially nuclear and wind exhibit different penetration patterns, the wind installed capacity accounts for more than 40 GW in 2050 in a new pricing regime, while the installed capacity of nuclear technology which is used as baseload plant drop to 1.2 GW. In the same vein, evidence of more capital investments is attributed to wind and gas technologies with £45 billion for wind technology in 2050, suggesting that the intermittent wind technology takes significant proportion of the electricity investment under the dual pricing regime as the wind technology receives more capital investments while the baseload nuclear technology witnesses capital disinvestments. However, the new pricing regime reduces the amount of the 2021 CO₂ emission level in the UK electricity generation system by 55% at 2050, compared to 80% reduction in the marginal pricing system.

The second simulation chapter, Chapter 4, extends the built agent-based model to quantify the amount of the hydrogen storage required to reduce the excess wind electricity waste in the UK electricity system. Furthermore, the chapter provides the economic evaluation of the hydrogen storage investment using the discounted cashflow method that allows for both market and policy uncertainties. By implementing these techniques, the chapter finds that the required hydrogen capacity increases more in the next 20 years when varying both carbon price and electricity price, compared to the fixed price case. Additionally, the minimum amount of hydrogen capacity to resolve this excess electricity waste, is above 2,000 MW with the associated investment cost of not less than £800 million. However, a significant reduction in the hydrogen capital cost decreases the investment cost of the hydrogen capacity, while more investment returns occur as the hydrogen lifespan increases. This indicates a trade-off between the hydrogen capital cost reduction and the lifespan expansion. In all the cases, the investment returns remain positive and robust, thus, suggesting that adequate and appropriate government incentives are needed to encourage private investment in the utility-scale hydrogen storage technology, which is economically viable, environmentally friendly, and socially worthwhile towards the UK net-zero targets. Also, the hydrogen policy needs to pay more attention on ensuring an optimal balance between the hydrogen capital cost reduction and the improved lifespan as both could influence the pace of the hydrogen storage investment in the UK electricity grid.