1. Introduction
Growing concerns about energy security, climate change, and energy shortage have aroused people’s awareness of increasing the proportion of low-carbon energy in total energy consumption, and the renewable energy sector has become one of the fastest-growing energy sectors. As of the end of 2018, the estimated share of renewable energy power generation in the total global power generation exceeded 26%, and renewable energy accounted for more than one-third of the global installed power generation capacity [
1]. The renewable energy sector has attracted remarkable investment, with more than
$280 billion invested in renewable power and fuels each year during 2015–2018, and with more renewable energy net capacity additions than fossil and nuclear energy combined [
1]. In addition, renewable energy sources will play an increasingly important role in meeting the world’s growing energy needs [
2]. According to the International Energy Agency’s (IEA) 2019 World Energy Outlook, under the Stated Policies Scenario, more than half of the increase in global energy demand by 2040 will be met by low-carbon energy sources [
3].
The rapid development and promising prospects of the renewable energy industry have stimulated the interest of private investors in renewable energy stocks. Research on the relationship between renewable energy stocks and other assets has grown in recent years, focusing on issues such as interdependence, causality, or spillover effect between renewable energy stocks, crude oil, or technology stocks. First, as crude oil plays an important role in global energy supply and substitutes with renewable energy, many studies have been conducted on the relationship between crude oil and renewable energy stocks. Henriques and Sadorsky, Kumar et al. and Ahmad et al., stated that rising oil prices have a positive impact on renewable energy stock prices [
4,
5,
6]. Reboredo and Ugolini and Song et al., studied the relationship between various fossil energies and renewable energy stocks, and found that crude oil price is the main factor affecting the price fluctuation of renewable energy [
7,
8]. Pham further analyzed the relationship between the stock performance of clean energy subsectors and the prices of crude oil, and they came to conclusion that the correlation between different sub-sectors and the price of crude oil varies. It is worth noting that some scholars have found that the correlations between crude oil prices and renewable energy stock prices or technology stocks have different features in different time periods or time scales. Managi and Okimoto introduced Markov-switching framework and found that there is a structural break in the system composed of crude oil, clean energy stocks and technology stocks at the end of 2007. Before the structural break, crude oil price had no significant impact on clean energy stocks, but the oil price shock after the structural change has a significant positive impact on renewable energy stocks [
9]. By employing Detrended Cross-Correlation Analysis (DCCA) framework, Paiva et al. found that crude oil and renewable energy are highly correlated only from mid-2008 to mid-2012 [
10]. Reboredo et al. used wavelet method to find that the interactions between crude oil prices and renewable energy stock prices are weak in the short term, and there is no linear Granger causality, but the dependence between the two increases in the longer time scale [
11].
Second, some scholars use different econometric methods and find that compared with the price of crude oil, there is a stronger correlation or spillover effect between technology stocks prices and renewable energy stock prices. Henriques and Sadorsky built a four-variable VAR model and found that the volatility of technology stocks has a greater impact on renewable energy stocks than that of crude oil prices [
4]. By constructing a multivariate GARCH model, Sadorsky found that the dynamic correlation between clean energy stocks and technology stocks is higher than that between clean energy and oil prices [
12]. Ahmad took more asset types into account based on Sadorsky’s research and came to a similar conclusion [
13]. Inchauspe et al., constructed an asset pricing model with time-varying parameters and found that the price returns of renewable energy stocks are much less affected by oil prices than those of technology stocks [
14]. Zhang and Du established a TVP-SV-VAR model to study the dynamic relationship between the stock prices of renewable energy, high-tech and fossil fuel companies, and found that renewable energy companies stock prices have a stronger correlation with high-tech stocks [
15]. Tiwari et al., investigated the tail dependence between crude oil, clean energy and technology stocks in different market scenarios and discovered that oil price volatility is not a key factor affecting the profitability of clean energy and technology companies, while clean energy and technology companies are strongly correlated [
16]. In addition, some scholars further find that there are differences in the correlations between crude oil, renewable energy stocks and technology stocks at different time scales. Ferrer et al., believed that the correlation between crude oil prices and renewable energy stock prices is low in both the short and long term, while there is a significant volatility spillover effect between technology stock prices and renewable energy stock prices in the short term [
17]. Through threshold cointegration test, Bondia et al., found that in the short run, alternative energy stocks are affected by technology stocks, oil prices, and interest rates; in the long run, alternative energy stocks are not affected by these factors [
18]. By constructing a cointegrating nonlinear auto-regressive distributed lag (NARDL) model, Kocaarslan and Soytas found that a rise in oil prices in the short term leads to an increase in clean energy investment, but in the long run, the price of crude oil has a negative impact on clean energy; in both the short and long term, there is a significant two-way positive effect between technology stocks and clean energy stocks [
19].
Third, carbon pricing policies aimed at mitigating climate change can also stimulate interest in renewable energy, but only a small number of studies have taken the carbon allowances price into account when exploring the volatility of renewable energy stock prices. Dutta et al., found that, by building the VAR-GARCH model, in Europe, the correlation between carbon allowances and the price return of clean energy stocks is not significant, but there is a significant volatility correlation between them [
20]. Xia et al., stated that electricity, coal or oil prices contribute more to changes in renewable energy earnings than carbon allowance prices [
21]. Lin and Chen, Jiang et al., took China’s carbon emission trading (CET) market as an example to study the correlation between carbon prices and renewable energy stock prices. Lin and Chen found that there is no significant volatility spillover effect between China’s carbon market and renewable energy stocks market, while Jiang et al., believed that renewable energy stocks have an impact on carbon allowance prices in the short and middle term [
22,
23].
Despite the growing body of research on the relationship between renewable energy stocks and assets such as crude oil, technology stocks or carbon allowances, there are still some research gaps. One of them is that few scholars pay attention to the interaction between renewable energy stocks and other assets in different frequency domains. However, given that different institutions differ in how they respond to and pay attention to information over different time horizons, we believe it is important to understand the frequency dynamics between renewable energy stocks and other assets. For market investors, due to different goals, preferences and risk tolerance, they often have different investment horizons. Specifically, short-term investors, such as day traders or hedge funds, will pay more attention to the short-term performance of the market, while long-term investors, such as large investment institutions, will more focus on the long-term performance of the market and adjust their investment strategies. Another research gap is that there are few studies on the relationship between carbon allowance prices and renewable energy stock prices. However, as the importance of global carbon emission reduction becomes increasingly prominent, it is of theoretical and practical significance to study the relationship between carbon allowance and renewable energy stocks. This paper introduces a recently developed connectedness approach to study the price return and price volatility among renewable energy stocks, global stock index, technology stocks, crude oil futures, carbon allowances, and the 10-year US Treasury note both in time domain and frequency domain.
The contributions of this research are mainly lie in the following four points:
First, this paper may be the first study to examine the spillover effect between renewable energy stocks, crude oil, technology stocks, and carbon allowances from a systematic perspective. To our knowledge, there are no studies on renewable energy spillovers that consider both technology stocks and carbon allowances. We believe it’s necessary to consider both assets because: (i) some studies show a stronger correlation or spillover effect between technology stocks and renewable energy stocks compared to crude oil; (ii) carbon pricing policies can stimulate interest in renewable energy, and some studies have shown that there is a positive correlation between carbon allowances and clean energy stocks.
Second, this paper uses a newly developed network connectedness approach to study the dynamic spillovers among renewable energy stocks and other assets over different investment horizons. Investors tend to choose different horizons due to their different objectives, preferences, and risk tolerance, yet there is a lack of research on the frequency dynamics between renewable energy stocks and other assets. The connectedness method proposed by Barunik and Krehlik, which adds the investigation on the frequency dynamics of connectedness on the basis of the connectedness network method proposed by Diebold and Yilmaz, and has been gradually applied to the fields of energy and finance in recent years [
17,
24,
25,
26]. Many existing studies on the relationship between renewable energy stocks and other assets use VAR and GARCH models and investigate the relationship between variables in the time domain through these types of econometric models, while this paper investigates the interactions between renewable energy stocks and other assets in both time and frequency domains.
Third, this paper studies both price return and price volatility spillovers, and research on these two channels can help investors and policy makers better understand the interactions between markets. In theory, renewable energy stocks can be linked to other financial markets, energy markets or carbon allowance markets through two ways. One is price return spillovers that occur during price discovery process, during which a price change in one market has an impact on the value of an asset in other markets, resulting in changes in their prices [
27,
28,
29]. The other is price volatility spillovers, in which the willingness of market participants to hold assets in other markets may change when one market fluctuates. To our knowledge, the spillovers of these assets under both channels have not been studied.
Fourth, this paper covers the period from the 2008 financial crisis to the 2021 COVID-19 pandemic, and separately studies the dynamic spillovers between renewable energy stocks and other assets during European debt crisis, oil price drop, and COVID-19 pandemic.
This paper can provide some implications to policy makers and investors for market or asset management, which will be conducive to maintain the stability of energy market or financial market, promote the development of renewable energy industry, and improve the risk management strategies of market investors. For policy makers, first, by studying the connectedness between the volatility of the renewable energy sector and the crude oil market at different scales, it is possible to assess the risk spillovers between markets and improve the price mechanism to prevent the market from violent turmoil. Second, by investigating the strength of the spillover effect between the returns of the renewable energy sector and crude oil and technology industries, it helps policy makers to evaluate the impact of other industries on the asset value of the renewable energy industry and to formulate corresponding policies to promote the development of the renewable energy industry. Third, studying the dynamic spillover effect between carbon allowances prices and the renewable energy sector is conducive to measuring the impact of carbon trading on the renewable energy sector in a quantitative and dynamic manner, and helps to assess the role of emission trading scheme in promoting the development of renewable energy. For market investors, the study of returns and volatility spillovers between various assets such as renewable energy stocks, crude oil, technology stocks and carbon allowances over different investment horizons will help investors to adjust the asset allocation of their portfolios in a timely manner. Investors can effectively reduce the market risk of investment by selecting assets with a certain hedging effect in an investment portfolio. In addition, portfolio management by investors is more difficult in times of turmoil in financial or energy markets. This paper analyzes the interactions between a large set of assets during the European debt crisis, the international oil price decline and the COVID-19 pandemic, which can provide a reference for investors in developing an asset diversification strategy. The remainder of the paper is organized as follows: The
Section 2 introduces the connectedness network approach proposed by Barunik and Krehlik; The
Section 3 describes the variables and makes a preliminary analysis of the data; In the
Section 4, we represent the empirical results of the connectedness analysis in time and frequency domain; The
Section 5 summarizes the main findings and makes implications for market investors and policy makers.
5. Conclusions and Discussion
The renewable energy sector has grown substantially over the past decades, which makes the renewable energy stocks become an investment asset attracting growing attention. Thus, the relationship between renewable energy stocks and other investment assets has aroused wide interest of scholars, policy makers and market investors. To obtain the frequency dynamics of spillovers between renewable energy stocks and other energy or financial assets, this paper employs the connectedness method proposed by Barunik and Krehlik (2018) to study the price return and price volatility connectedness among renewable energy stocks, global stock index, technology stocks, crude oil futures, carbon allowances, and the 10-year US Treasury note both in time and frequency domains. Through empirical research, this paper has the following main findings:
First, in the system composed of renewable energy stocks, technology stocks, crude oil futures, and carbon allowances, both price return spillovers and price volatility spillovers are high and are mainly driven by short-term (within one week) spillovers. This shows that the information about price return and price volatility in the system is transmitted quickly, and market participants process the information efficiently. On a longer time horizon (more than one week), the markets in the system may be mainly affected by its own fundamentals, rather than by prices of other markets in the system. Scholars such as Wang and Wang, Cui et al., and Umar et al., who also adopted the approach proposed by Barunik and Krehlik (2018) to study the volatility spillover in the system consisting of energy and financial markets, have different views on whether the overall risk spillover occurs mainly in the short run or in the long run [
24,
52,
53]. We believe that this may be related to the type of energy sources selected and the different regions studied. However, these studies all point out that the overall volatility spillover level of the energy market or financial market increases significantly during periods of economic or financial market turbulence, which is consistent with the findings of this paper.
Second, we find something new about renewable energy’s relationship with crude oil or technology stocks, respectively. We find that we cannot simply generalize to which asset renewable energy stocks are more strongly associated with but should instead consider the frequency dynamics of the relationship between assets. Our study in the time domain has similar finding to that of Sadorsky, Ahmad, Inchauspe et al. in that technology stocks have a more significant impact on renewable energy stocks compared to crude oil [
12,
13,
14]. However, through the study in the frequency domain, we find that renewable energy stocks show a more complex relationship with the two at different time scales.
- i.
In the price return system, only in the short term, renewable energy has a significant spillover effect on the price of crude oil. In longer time scales, the impact of crude oil on renewable energy stocks is enhanced. In the price volatility system, crude oil has certain influence on the risk of renewable energy stock market in all frequency bands, which means that renewable energy stock investors need to pay attention to the volatility of crude oil prices and adjust the asset allocation in their portfolios in a timely manner. In addition, policy makers should improve the price mechanism of the crude oil market to prevent the violent fluctuation of crude oil prices from having a huge impact on the renewable energy financial market.
- ii.
In the price return system, renewable energy stocks have a significant spillover effect on technology stocks in the short term, while technology stocks have a significant spillover effect on the renewable energy over a time scale of more than a week. This may mean that the long-term return of renewable energy is more strongly influenced by the return of the technology industry, and it’s an effective way to achieve sustainable development of the renewable energy industry by promoting the progress of science and technology. In the price volatility system, only in the short term, technology stocks have a significant impact on renewable energy stocks. In longer time scales, the dominant role of technology stocks diminishes. For short-term investors, including technology stocks and renewable energy stocks in the same portfolio may increase investment risk.
Third, only in the short term, renewable energy stock prices have significant price return spillover effect and price volatility spillover effect on carbon allowance prices. In longer terms, renewable energy stock prices have limited influence on carbon allowances prices while other factors such as energy prices, climate and policies may have a greater impact. Therefore, short-term investors in carbon markets should pay more attention to renewable energy stocks, while other investors may need to focus more on other energy prices, climate, or policies. In addition, over a time scale of more than a week, the carbon price has a certain impact on the return and risk of renewable energy stocks.
There are few studies on the relationship between the price of carbon allowances and the volatility of renewable energy stock prices, such as those by Dutta et al., Xia et al., Lin and Chen and Jiang et al., among which only Jiang et al. consider the time scale factor. Different from the wavelet method used by Jiang et al., by introducing a frequency-dependent connectedness network model, we not only consider different time scales, but also study the dynamic relationship between carbon prices and renewable energy stock prices.
Fourth, during the turmoil in the financial market, macro economy or energy market, the price return and price volatility spillover effect of the system composed of renewable energy stocks and other assets will become stronger, especially the volatility spillover effect. This is consistent with previous finding that major crisis events enhance the risk transmission between energy and financial markets [
24,
30,
52,
53]. During the European debt crisis, the international oil price decline and the COVID-19 pandemic, the total volatility spillover of the system has increased substantially, which is not caused by spillovers within one week, but by spillovers in longer time scales. This suggests that risk transmission between renewable energy stocks and other assets has been profoundly affected.
This research has made new findings on the relationship between the renewable energy and investment assets such as technology stocks, crude oil futures and carbon allowances, which can not only provide reference for investors to manage risks and optimize asset allocation strategies, but also provide suggestions for policy makers to promote the development of renewable energy industry and maintain market stability. The limitation of this paper is that it only analyzes the spillover results between renewable energy stocks and other investment assets in different frequency domains, and lacks research on the hedging ratio between different assets, which will be the focus of our further research.