Global climate change necessitates innovative solutions for carbon emission reduction. While fossil fuels drive economic development, their reliance exacerbates energy shortages and global warming. The Paris Agreement and subsequent net-zero emission targets underscore the urgency for collective action, including technological innovation. China's commitment to peak carbon emissions by 2030 and achieve carbon neutrality by 2060 highlights the significance of effective carbon reduction policies. Existing literature explores the impacts of carbon taxes, trading schemes, and caps on carbon reduction and the macroeconomy. Studies reveal potential industrial relocation due to differing carbon policies, posing challenges for China's transition to a green growth model. While carbon taxes incentivize emission reductions, they may negatively impact social welfare. Carbon trading systems, however, appear more effective in reducing emissions and stimulating technological innovation. The synergistic effect of combining carbon trading with carbon caps is also recognized. Directly subsidizing low-carbon energy technologies is another crucial policy measure. However, the literature on energy subsidy policies remains relatively underdeveloped, with conflicting findings on their long-term and short-term effects and varying subsidy intensities. Existing research lacks detailed exploration of the mechanisms through which energy technology subsidies influence technological innovation, energy structures, and macroeconomic output. A gap also exists in comparative analyses of different policy approaches and their synergistic effects. Technological innovation is pivotal for mitigating carbon emissions. Advances in energy technologies improve production efficiency and reduce energy intensity. Fossil fuel technologies, such as CCUS, improve efficiency and emission intensity, while renewable energy technology advancements directly displace high-emission fossil fuels. Existing studies often focus on individual energy technologies, overlooking the interaction between fossil fuel and renewable energy technologies and their combined effects. Understanding macroeconomic oscillations induced by carbon reduction policies is crucial for effective policy formulation. While decomposition analysis and statistical models (IPAT, KAYA, STIRPAT) provide insights, they are limited by the Lucas Critique. Integrated Assessment Models (IAMs) and Computable General Equilibrium (CGE) models are used for policy assessment but often ignore uncertainty. DSGE models, however, incorporate intertemporal decisions and uncertainties, offering a more robust approach to analyzing policy impacts. Existing DSGE studies often overlook the heterogeneous impacts of different production sectors and the role of energy technology innovation. This study bridges these gaps by constructing a multi-technology sectoral NK-DSGE model to compare the effects of carbon emission caps, fossil fuel technology subsidies, and renewable energy technology subsidies in China's context. This model incorporates both fossil fuel and renewable energy technology sectors, addresses the under-researched area of energy technology subsidies, and offers a comparative analysis of different policy approaches.

The literature on carbon reduction policies reveals a variety of approaches and their associated effects. Carbon taxes, while incentivizing emissions reductions, can negatively impact social welfare if not carefully designed. Carbon trading schemes, often coupled with carbon caps, appear to be more effective in stimulating technological innovation and achieving emission reductions. However, the literature examining the effects of energy technology subsidies is relatively less developed, with conflicting findings regarding their short-term and long-term macroeconomic impacts. Some studies highlight the positive effects of new energy technology subsidies on energy structure transitions and economic growth, while others indicate potentially negative macroeconomic consequences. This discrepancy likely stems from differences in the long-term versus short-term effects, as well as varying subsidy intensities. There is a lack of comprehensive theoretical understanding of how energy subsidies directly or indirectly impact technological innovation, energy structures, and macroeconomic outputs. Furthermore, existing studies often examine individual policies in isolation, overlooking the potential synergistic effects of combining different policy instruments. This paper directly addresses these gaps in the literature by providing a comprehensive comparative analysis of multiple policy scenarios.

The study employs a multi-technology sectoral New Keynesian Dynamic Stochastic General Equilibrium (NK-DSGE) model. This model incorporates six sectors: households, intermediate goods, final goods, fossil fuel technology, renewable energy technology, and government. The model framework, illustrated in Figure 1, shows material and fund flows between these sectors. The household sector owns factors of production (labor, capital, fossil fuel, renewable energy), supplying them to production sectors and receiving income. Energy technologies, produced by the fossil fuel and renewable energy technology sectors, are crucial inputs for the intermediate goods sector. Intermediate goods are supplied to the final goods sector, which in turn supplies to households and pays profits. The government collects taxes, purchases final goods, and provides transfer payments. Pollutant emissions, energy productivity, and carbon reduction policies are integrated into the model. A key distinction from prior studies is the inclusion of separate fossil fuel and renewable energy technology sectors, allowing for an analysis of their heterogeneous effects on emission reduction and economic development. The model incorporates key equations governing household utility maximization, intertemporal budget constraints, investment adjustment costs (GQ), and price adjustments (Calvo). The energy technology sectors are modeled with Cobb-Douglas production functions, incorporating capital and labor inputs. The model includes energy technology research and development efficiency, following an AR(1) process. The intermediate goods sector's production function is also Cobb-Douglas, incorporating capital, labor, fossil fuel, and renewable energy inputs. Energy efficiency is linked to energy technology inputs, utilizing a 'learning by doing' (LBD) approach. The model includes equations for emission reduction rates, pollutant emissions, emission reductions, and emission reduction costs. The effect of pollutant stock on labor efficiency is also incorporated. The final goods sector uses a CES technology for production. The government's budget constraint includes taxes, emission permit fees, and transfer payments. The model also incorporates price dispersion and market clearing conditions. The model parameters were calibrated using a combination of existing research and Chinese economic data. Bayesian estimation was used to estimate parameters related to AR(1) processes for economic shocks and policy shocks using quarterly data from China from 2000Q1 to 2020Q4. The data included total output, government expenditure, fossil fuel technology, and renewable energy technology, which were processed using various techniques (deseasonalization, logarithmic transformation, HP filter) to extract volatile components.

The study establishes four scenarios: Business-as-usual (BAU), Carbon Emission Cap (CEC), Fossil Fuel Technology Subsidy (FTS), and Renewable Energy Technology Subsidy (RTS). The long-run steady-state analysis reveals that implementing carbon reduction policies negatively impacts consumption, but positively impacts labor, capital, and renewable energy inputs, leading to increased total output. All three policy scenarios demonstrate significant reductions in pollutant emissions and pollutant stock compared to the BAU scenario. The FTS and RTS scenarios show even greater emission reductions than the CEC scenario. The analysis of exogenous shocks reveals distinct responses across scenarios. A positive Total Factor Productivity (TFP) shock increases output, consumption, investment, and capital stock across all scenarios, but the BAU scenario shows the most pronounced fluctuations. The demand for renewable energy technologies initially decreases, while the demand for fossil fuel technologies increases, due to the existing energy structure. However, pollutant emissions increase in all but the CEC scenario. Government expenditure shocks similarly stimulate macroeconomic growth, but reduce investment and capital stock. Renewable energy technology demand increases, while fossil fuel technology demand decreases. Pollutant emissions increase in BAU, FTS, and RTS scenarios. The Energy Technology Research Productivity shock unexpectedly decreases total output due to resource reallocation. Demand for renewable energy increases, while the demand for fossil fuel technology initially increases before declining. Pollutant emissions decrease. A positive energy efficiency shock increases output, consumption, investment, capital, and labor. Demand for fossil fuel technology increases initially, but demand for renewable energy technology initially decreases. Pollutant emissions increase in all but the CEC scenario. The analysis of carbon emission reduction policy shocks reveals that pollutant emissions are pro-cyclical under the carbon emission cap policy but counter-cyclical under energy technology subsidy policies. The economic variables respond more sharply to shocks under the carbon emission cap policy than under the energy technology subsidy policies. Energy technology subsidies are more effective in reducing macroeconomic volatility.

The findings address the research question by comparing the effectiveness of different carbon reduction policies in China. The results demonstrate that, in the long run, energy technology subsidy policies are more effective in reducing carbon emissions compared to the carbon emission cap policy. The higher effectiveness of the subsidy policies is attributed to their ability to directly incentivize technological innovation and drive changes in energy structures. Moreover, the analysis of exogenous shocks highlights the importance of considering the dynamic interplay between economic growth and environmental sustainability. The pro-cyclical nature of emissions under carbon caps contrasts sharply with the counter-cyclical behavior observed under subsidy policies. The latter suggests that subsidies can effectively decouple economic growth from emissions increases. The study's significance lies in its comprehensive analysis of multiple policy approaches using a sophisticated DSGE model, incorporating sector-specific details and uncertainties. The findings are relevant to policymakers in China and other countries striving to achieve carbon neutrality goals. They emphasize the need to consider not just the long-term effects of carbon reduction policies, but also their short-term macroeconomic and environmental consequences. The results suggest that a phased approach, combining carbon caps with targeted energy technology subsidies, could be an effective strategy for achieving both emission reduction and economic growth.

This research contributes to the literature by providing a comprehensive analysis of carbon reduction policies using a multi-technology sectoral DSGE model. The findings suggest that energy technology subsidy policies are more effective for long-term emission reduction than carbon emission cap policies. The study also highlights the importance of considering both short-term and long-term effects of different policies and the need for a phased approach that combines carbon caps with targeted subsidies. Future research should investigate heterogeneous household structures, the interaction between carbon reduction policies and other economic and environmental policies, and extend the model to an open economy framework to account for global trade dynamics.

The study's limitations include the model's assumptions, such as the representation of households as homogenous and the focus on China's economy without considering international trade effects. The model's calibration and Bayesian estimation rely on specific data and methodologies, which may limit the generalizability of the findings. The lack of detailed consideration of specific technological pathways and their uncertainties is another potential limitation. Further, the model does not capture the full complexity of political and social factors that can influence policy implementation and effectiveness.