Small fish in big ponds : Connections of green finance assets to commodity and sectoral stock markets

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Small fish in big ponds : Connections of green finance assets to commodity and sectoral stock markets © 2022 the Authors Published version Naeem, Muhammad Abubakr; Karim, Sitara; Uddin, Gazi Salah; Junttila, Juha Naeem, M. A., Karim, S., Uddin, G. S., & Junttila, J. (2022). Small fish in big ponds : Connections of green finance assets to commodity and sectoral stock markets. International Review of Financial Analysis, 83, Article 102283. https://doi.org/10.1016/j.irfa.2022.102283 2022 International Review of Financial Analysis 83 (2022) 102283 Available online 2 July 2022 1057-5219/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Small fish in big ponds: Connections of green finance assets to commodity and sectoral stock markets Muhammad Abubakr Naeem a , b , Sitara Karim c , Gazi Salah Uddin d , 1 , Juha Junttila e , * a Accounting and Finance Department, United Arab Emirates University, P.O. Box 15551, Al-Ain, United Arab Emirates b South Ural State University, Lenin Prospect 76, Chelyabinsk 454080, Russian Federation c Nottingham University Business School, University of Nottingham Malaysia Campus, Semenyih, Malaysia d Department of Management and Engineering, Link¨ oping University, Link¨ oping, Sweden e University of Jyv¨ askyl¨ a School of Business and Economics, Jyv¨ askyl¨ a, Finland ARTICLE INFO JEL classifications: C22 C58 G01 G11 G15 Keywords: Green markets US sectors Commodities Connectedness Time-frequency ABSTRACT We analyze return and volatility connectedness of the rising green asset and the well-established US industry stock and commodity markets from September 2010 to July 2021. We find that the time-varying return and volatility connectedness have exhibited serious crisis jumps. Some individual assets of both the green and commodity markets are in connection to the US sectoral stock market returns, and the volatility connections are even more common than the return connections. Furthermore, some financial and economic uncertainty indicators manifest positive impacts from the volatility of some ‘big pond’ markets for e.g. commodities, whereas some others affect the connectedness negatively. Additional analysis of financial and economic uncertainty indicators manifests positive impacts from the volatility of some ‘big pond’ markets, e.g., commodities, while others negatively affect the connectedness. 1. Introduction Sharing similar features with corporate treasury investments, green investments are new forms of financial intermediation whose proceeds are directly attributed toward environment-friendly and climateoriented projects to reduce carbon emissions and encourage adaptation of renewable energy sources. In the future, there will be even more increasing attention of policymakers, regulation bodies, and investors concerning the magnified benefits of green markets as they are only weakly correlated with other markets (Pham & Huynh, 2020). Hence, from a practical point of view, they might also offer diversification benefits to the investors. At the first steps, introduced by the European Investment Bank in 2007, green investments were at the epicentre of financial regulators to channel the investments and assets to achieve sustainability and effectively tackle environmental challenges. Prior literature reveals green markets as an operative means to finance climate-oriented projects to achieve a low-carbon economy (Andersen, Bhattacharya, & Liu, 2020; Leitao, Ferreira, & Santibanez-Gonzalez, 2021). Since the enthusiasm about green investments has started to outperform nowadays that of traditional financial assets, several stock exchanges across the world are introducing specialized investments which fulfil the green objectives of the investors, and a sharp increase in the allocations to green investments has been already reported, going from $11 billion to $350 billion between 2013 and 2020 (Climate Bonds Initiative, H1–, 2020). The concentration of regulatory bodies on the COP26 accord and Paris Agreement from 2015 shows sustained pressure by governments to overcome climate degradation by reducing global warming below 2–1.5 degrees. Following this, green markets have exhibited >100% annual growth rate and they are expected to account for one-third of global assets by 2025. Hence, it is also evident that these markets might provide number of useful benefits in e.g., managing risk and reducing the losses on investments under extreme circumstances * Corresponding author at: University of Jyv¨ askyl¨ a School of Business and Economics, University of Jyv¨ askyl¨ a, PO Box 35, FI-40014, Finland. E-mail addresses: [email protected] (M.A. Naeem), [email protected] (S. Karim), [email protected] (G.S. Uddin), juha-pekka. [email protected] (J. Junttila). 1 Part of this research was conducted when Gazi Salah Uddin was visiting the University of Jyv¨ askyl¨ a School of Business and Economics under the JYU Visiting Fellow Programme (Grant number 444/13.00.05.00/2021) Contents lists available at ScienceDirect International Review of Financial Analysis journal homepage: www.elsevier.com/locate/irfa https://doi.org/10.1016/j.irfa.2022.102283 Received 2 February 2022; Received in revised form 21 April 2022; Accepted 28 June 2022 International Review of Financial Analysis 83 (2022) 102283 2 (Karim, Lucey, & Naeem, 2022; Karim, Lucey, Naeem, & Uddin, 2022; Karim & Naeem, 2021; Karim & Naeem, 2022; Naeem & Karim, 2021). Based on this development, the relationships between green markets, US sectoral markets, and commodity markets are intuitively appealing as these market segments represent three different financial markets with unique underlying characteristics to absorb shocks and respond to the market uncertainties (Kilian & Zhou, 2018; Marshall, Nguyen, Nguyen, Visaltanachotin, & Young, 2021). Following the cross-market diversification perspective, Ngene (2021) has elaborated that industrylevel diversification will dominate portfolio diversification in the stock markets in the long run, specifically in North America and Europe. Importantly, domestic investors overweigh the domestic investments compared to international investments and forgo the risk-adjusted diversification returns. As the US sectors are not sheltered from the market ups and downs, investors reallocate their investments during bullish and bearish market conditions to improve the risk-adjusted returns in times of economic recession. Similarly, the financialization of commodities provides an attractive hedging tool due to their risk mitigation ability during abrupt economic swings (Balli, Naeem, Shahzad, & de Bruin, 2019; Prokopczuk, Stancu, & Symeonidis, 2019). As the pricing mechanisms of commodities might differ from the conventional asset classes, for example, the traditional (spot/futures market) demand and supply shocks substantially determine the commodity prices. For example, the demand for commodities is associated with cumulative global aggregate demand conditions (Bakas & Triantafyllou, 2018), except for perhaps, e.g., precious metals as they have often been considered to provide hedging facilities in uncertain times. Following this, volatility in the macroeconomic conditions and unexpected frequent downfalls might result in catastrophic consequences for risk-sensitive investments (Apostolakis, Floros, Gkillas, & Wohar, 2021; Wang, Xie, Zhao, & Jiang, 2018). Previous literature has postulated several uneven circumstances where growth in the financial markets, commodities, various sectors and industries is seized (Abakah, Addo, Gil-Alana, & Tiwari, 2021; Mensi, Nekhili, Vo, Suleman, & Kang, 2021; Naeem, Adekoya, & Oliyide, 2021; Naeem, Rabbani, Karim, & Billah, 2021). The most recent global outbreak resulted in a 5% decline in the US real GDP from the last quarter of 2019 to 2020 due to the pandemic’s unprecedented havoc. Moreover, the unemployment rate erupted from 4.4% in March 2020 to 14.7% in April 2020. The crash of numerous stock and equity indices, and the fall in the valuation of banks and financial institutions by 39% in the US restricted traveling, and the increased loss exposure to the oil, energy and gas sector represents the extreme costs and damage caused by the COVID-19 pandemic (Shahzad, Bouri, Kristoufek, & Saeed, 2021). Meanwhile, financial markets experienced endangered susceptibility to the unexpected shocks propelled out of this world health emergency. Therefore, these shocks are central to the interconnectedness and volatility spillovers of financial assets, as more intensive risk spillovers spike the correlations among the markets (Kang, Hernandez, Sadorsky, & McIver, 2021). Given these turmoil periods, green investments seem to have remarkably sheltered the other investments from volatile and distressing episodes, accelerated the cross-market financialization, and provided noticeable evidence of diversification and hedging facilities (Arif, Hasan, Alawi, & Naeem, 2021; Naeem, Adekoya, & Oliyide, 2021; Naeem & Karim, 2021; Reboredo, Ugolini, & Aiube, 2020). Based on this practical observation, a thorough examination of the return and volatility connectedness between green and conventional assets presents a new dimension of analysis useful for the investors, policymakers, and strategists for designing their portfolios, devising useful policies and implementing them to reap the benefits of adding diversifiers in their asset-mix. Theoretically, Modern Portfolio Theory (MPT) offers useful insights to choose among different market segments for risk mitigation and diversification objectives of the investors (Markowitz, 1952). In addition, we argue that several economic and financial factors influence the spillover network of the markets. This is consistent with market reaction hypothesis where investors not necessarily rely on all the available information, but rather their decisions are heterogenous and vary with the changing financial and economic circumstances (Naeem, Farid, Qureshi, & Taghizadeh-Hesary, 2022). Prior literature, for instance, Elsayed et al., (2022), Khalfaoui et al., (2022), Mensi et al., (2022), Tiwari et al., (2022), and Urom et al., (2021) present limited intuitions to the current body of knowledge by empirically testing the dependence of green markets from the other financial market segments using a multitude of econometric techniques. However, we differ from these studies by exhibiting unique return and volatility connectedness networks among green bonds (small fish), US sectors, and commodities (big ponds) based on employing diverse econometric techniques. In light of this background, the current study contributes to the existing literature in many ways. First, this is the pioneer study that investigates the connectedness and spillover network among green markets (small fish), US sectors, and commodities (big ponds) to inspect their underlying relationships based on the utilized return and volatility connectedness measures. The rationale behind using both return and volatility connectedness measures implies that markets behave differently under the conditions of average returns and uncertain circumstances (Umar, Adekoya, Oliyide, & Gubareva, 2021). Thus, the focus on return and volatility connectedness provides a comprehensive outlook for assessing the variations in spillovers during stable and turbulent time periods. Second, we employed the time-frequency approaches of Diebold & Yilmaz (2012, Diebold and Yılmaz, 2014) and Baruník and Kˇ rehlík (2018) to compute the spillover network. The time-based connectedness approach of Diebold and Yilmaz (2012) addresses the network connectedness of markets, while the frequency-based approach sufficiently describes the spillovers given various time horizons. Third, as additional evidence, the study further examines the impact of several economic and financial uncertainties on the return and volatility connectedness of green markets, US sectors, and commodities. Fourth, theoretically, the study embraces support from modern portfolio theory for indicating the investors’ choices to offset risks of their investments and achieve diversification benefits. Moreover, to unveil the impact of several economic and financial uncertainties on the return and volatility spillovers, we posit that investors’ reaction to various market conditions are heterogeneous. Thus, in line with market reaction hypothesis, we test the influence of financial and economic factors on the spillover networks. Finally, the study brings intriguing findings for policymakers, regulation authorities, financial market participants, investors, and portfolio managers to diversify their portfolios using green markets (small fish) and overcome the risk of other big ponds investments. Our findings reveal complex intra-group and moderate inter-group return and volatility connectedness. The US stock market sectors reveal the strongest intra-group return and volatility connectedness. Alternatively, the system-wide connectedness for total, short- and longterm horizons indicates high spillover from green markets and commodities to the US sectors. Time-varying attributes of return and volatility connectedness exhibit that the markets have suffered from shocks during unexpected economic periods such as the European Debt Crisis, Shale oil crisis, Chinese stock market crash, and COVID-19 pandemic. Moreover, short-run return spillovers dominated the long-run spillovers, reflecting financial contagion during economically fragile times. Furthermore, time-varying NET return and volatility connectedness characterized the US sectors as the NET transmitters of spillovers, whereas the green markets and commodities are NET receivers of return and volatility spillovers with significant time-varying features. Investigating the financial and economic drivers of connectedness among green markets, US stock market sectors, and commodity markets highlights the positive driving effect of implied volatilities of Russel Index and Gold and the negative driving effect of implied volatilities of exchange rates and bond markets. Moreover, we report a positive effect of the UK economic policy uncertainty and infectious disease tracker for the underlying connectedness in return and volatilities of green markets, US sectors, and commodities. Based on our results, we propose some M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 3 new implications for policymakers, investors, financial market participants, and portfolio managers trying to diversify their portfolio risks and derive appropriate strategies from short and long-run perspectives. The remaining parts of the paper are structured as follows: Section 2 reviews the earlier empirical studies; Section 3 elaborates methodology and data; Section 4 presents empirical results and discussion; and finally, Section 5 concludes the study with our main policy implications. 2. Literature review and theoretical background 2.1. Theoretical background Modern portfolio theory (MPT) proposed by Markowitz (1952), offers strong theoretical underpinnings for the construction of diversified portfolios in order obtain the expected returns in connection to a certain level of market risk. At the same time, diversification cannot completely eliminate the risk of an investment, but rather, the diversification provides avenues for optimizing the investment streams and achieving higher expected returns. Hence, the essentials of MPT include the quantification of the risk-return relationship embraced from portfolio management theory (Omisore, Yusuf, & Christopher, 2012). MPT mathematically quantifies the diversification objectives of investors with the aim of selecting a set of investment avenues that bear relatively lower risk than an individual asset. Based on these ideas, we claim that nowadays the green markets, despite of begin still an emerging market segment in terms of their financial integration (funding) role, might offer greater diversification possibilities against other, more established financial market segments (Elsayed, Naifar, Nasreen, & Tiwari, 2022; Khalfaoui, Jabeur, & Dogan, 2022). Hence, we first hypothesize that: H1. : Green markets offer new diversification benefits to mitigate the risk of US sectoral stock markets and commodity markets. In addition, the market reaction hypothesis contends that investors, before making their investment decisions, do not always rely on the readily available information. Instead, their attitudes toward various investment streams determine the heterogenous responses across multiple financial markets (Naeem, Karim, Jamasb, & Nepal, 2022; Naeem, Pham, Senthilkumar, & Karim, 2022). In these circumstances, it is imperative to identify the impact of numerous financial and economic uncertainties on the spillover networks formed among green markets, US sectors, and commodities. We further extend our argumentation by suggesting that nonlinearities exist among financial markets embarked with structural variations. Therefore, investors must consider the impact of financial and economic uncertainties before reaching their investment decisions. Building on these arguments, we frame our second hypothesis in the form: H2. : Financial and economic uncertainties significantly drive the spillover network of green markets, US sectors and commodities. 2.2. Earlier empirical literature The existing strand of literature has most often focused on the advantages of green bonds for various investors, policymakers and regulatory authorities. Tang and Zhang (2020) reported that the issuance of green bonds has positively impacted the stock market indices. Likewise, Russo, Mariani, and Caragnano (2021) investigated the determinants of green bond performance for developing sustainable strategies. Another part of the literature has concentrated on the similar characteristics of green bonds and conventional markets with other financial assets (Ferrer, Shahzad, & Soriano, 2021). On the other hand, several studies have studied the hedge and safe haven characteristics of green bonds against several commodities, bonds, and other financial market segments (Arif, Naeem, Farid, Nepal, & Jamasb, 2021; Naeem, Adekoya, & Oliyide, 2021) and suggested that green bonds act as a potential diversifier during normal economic conditions, but they offer safe-haven attributes during the crisis periods. Meanwhile, several studies have reported mixed evidence for the connectedness structure of green bonds with other markets (Nguyen, Naeem, Balli, Balli, & Vo, 2020; Reboredo et al., 2020). The previous literature presents evidence about the connections between green and conventional markets using various methodologies, such as comovement analysis based on wavelets (Mensi, Naeem, Vo, & Kang, 2022; Mensi, Rehman, & Vo, 2020; Nguyen et al., 2020), asymmetric time-frequency connectedness (Naeem, Adekoya, & Oliyide, 2021), quantile connectedness for estimating return connectedness of clean and dirty energy investments (Saeed, Bouri, & Alsulami, 2021; Tiwari, Abakah, Gabauer, & Dwumfour, 2022; Urom, Mzoughi, Abid, & Brahim, 2021), extreme quantile approach (Naeem, Rabbani, et al., 2021), time-varying optimal copula approach (Naeem & Karim, 2021), Diebold-Yilmaz framework (Bahloul & Khemakhem, 2021; Zhao, Umar, & Vo, 2021), and cross-quantilogram (Arif, Hasan, et al., 2021). However, the literature examining the determinants of a given or observed relationship is limited. For instance, Balli, Hasan, Ozer-Balli, and Gregory-Allen (2021) investigated the role of US uncertainty in driving spillovers of the global stock market. The authors reported that global factors, such as US uncertainties, substantially drive the US spillovers to global stock markets. Moreover, Abbas, Hammoudeh, Shahzad, Wang, and Wei (2019) reported that macroeconomic variables drive the connectedness of G7 markets. Finally, from the funding perspective of firms in general, it is worth to mention here also that some of the papers in the most recent literature propose that actually the ‘green finance certification’ allows managers to signal firms’ efficiency at addressing for example the energy transition activities. For example, the paper by Daubanes, Shema, and Rochet (2022) proposes that the firm-level green bond issuance signals firm’s amplified incentives to decarbonize its production, too. Their theoretical model predicts that firms’ managers are more inclined to issue green bonds when they are more interested in stock prices, and hence, the stock market valuation of their firm. They also test this prediction empirically by exploiting cross-industry differences in the stock-price sensitivity of managers’ compensation and cross-country variations in effective carbon prices, finding that the effect of managers’ incentives on green bond issuance increases with carbon penalties. This suggests that green bonds are complements to, rather than substitutes for, carbon pricing. Hence, based on this paper, from an investor’s point of view, it is obviously relevant to analyze also the market level return and volatility connections of wider set of green assets (than just green bonds) on the US sectoral stock markets, and also some relevant green transition related commodity markets. This is the focus of our empirical analysis, and next we describe the contents of it in more details. 3. Methodology and data This study investigates the return and volatility connectedness of green asset markets, US stock market sectors, and commodity markets. First of all, for estimation purposes, all the analyzed asset price series (P) are converted into log returns (R, in %) based on the first differences of prices: Rt=ln(Pt−Pt−1)×100 (1) where t refers to the time period of observation. 3.1. Volatility estimation For computing volatility, the individual time series of returns belonging to the vector of return series R t =[R 1t ,…….,R nt ]′is supposed to be given as an AR(1) process as follows: Rt= μ t+γRt−1+ ε t(2) The vector of constant terms is denoted by μ , whereas ε t =[e 1t , M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 4 ……, ε nt ] illustrates the vector of error terms. In addition, the conditional volatilities h it 2 (for each asset i) are based on estimating the univariate GARCH (1,1) models 2 for each of them given as: h2 it = ω +a ε 2 it−1+βh2 it−1(3) where ω >0, α ≥0, and β ≥0, and α +β <1. 3.2. Connectedness approach Based on Diebold and Yılmaz (2014), this study employs a unique connectedness measure derived from the variance decomposition matrix of the vector autoregressive (VAR) technique. The stationary N-variable vector is denoted as VAR(p), so for each component of the N-variable vector we have y t =∑ i=1 p ω i y t−i + ε t , where ε t ~(0,Σ) whereas the moving average representation is given by y t =∑ i=0 ∞ ∅ i ε t−1 and ∅ i is the combination of N ×N coefficient matrices obeying the recursion ∅ i = ω i ∅ i−1 + ω i ∅ i−2 +… + ω p ∅ p−1 and ∅ 0 represents identity matrix where ∅ i = 0 and i ≪ 0. In this way, moving averages facilitate understanding the dynamics. For this reason, variance decompositions are employed to obtain the transformations through moving averages. It also splits variable error variances into parts through H-step-ahead forecast, denoted as various shocks in the system. Orthogonality in the variables is attained using the Cholesky factorization, which ascertains the ordering of the variables. Following Pesaran and Shin (1998), the generalized approach is employed, allowing appropriate treatment of the correlated shocks. Thus, entries of the connectedness table are denoted as c ij g(H) which measure the contribution of variable j to H-step-ahead generalized forecast error variance of variable i given as: cg(H) ij = σ −1 jj ∑H−1 h=0(e′ i∅h∑ej)2 ∑H−1 h=0(e′ i∅h∑∅′ hej)2(4) Here the non-orthogonalized VAR representation of the covariance matrix is represented by Σ. Furthermore, σ jj is the standard deviation of the j-th diagonal component. For the i-th component, the selection vector e i has value of 1 and 0 otherwise. ∅ h represents the coefficient matrix of non-orthogonalized VAR model, which multiplies h-lagged errors in infinite moving averages. In the connectedness table, c ij g(H) estimated pairwise directional connectedness from j to i are calculated based on CH i←j=cg(H) ij (5) On the other hand, the total directional connectedness from others to i in the off-diagonal sum of rows is represented as: CH i←•=∑ N j=1 j∕=i cg(H) ij (6) However, the total directional connectedness to others from j in the off-diagonal sum of columns is represented as: CH •←j=∑ N i=1 i∕=j cg(H) ij (7) Finally, the system-wide total connectedness is obtained by summing the to-others and from-others elements of variance decompositions matrix given as: CH=1 N∑N i,j=1 i∕= j cg(H) ij (8) The structural connectedness table is graphically visualized so that the individual markets such as green markets, US stock market sectors and commodity markets represent nodes and the arrows present pairwise connectedness among these markets. 3.3. Decomposing frequency connectedness In this step, frequency connectedness is decomposed into short- and long-run frequencies by taking into account several spectral representations of variance decompositions. Apart from shock impulses, these decompositions are based on frequency responses to shocks. Thus, the frequency response function is denoted as ℵ(e −i ω g ) =∑ g e −i ω g ℵ g , obtained as Fourier transform of the coefficients ℵ g with i= −1 √. In this way, the Fourier transform for moving averages MA(∞) for spectral density of UV t at frequency ω is filtered as: SUV ( ω ) = ∑ ∞ g=− ∞ E(UVtUV′t−g)e−iwg = ℵ(e−iw)∑ℵ′(e+iw)(9) Here S UV ( ω ) denotes the key quantity power spectrum and frequency dynamics rely on this function as it describes the distribution of variance over frequency components ω . Nonetheless, the frequency domains are explained by the spectral decomposition of covariance in the form of E (UV t UV t−g ′) =∫ −φ φ S γ ( ω )e i ω g d ω . Baruník and Kˇ rehlík (2018) explained the comprehensive derivation of quantities, whereas we describe the connectedness of markets at varying frequencies. Therefore, spectral quantities are transformed by standard Fourier transforms across interval’s cross-spectral density d =(a,b) : a, b ϵ (−φ,φ), a ≪ b as: ∑ ω  ℵ( ω )∑ ℵ′( ω )(10) For ω ∈{aG/2 π ,…,bG/2 π }where  ℵ( ω ) = ∑ G−1 g=0 ℵge−2iφ ω /G(11) And ∑ ε ′ ε /(T−x)where x is the correction for loss of degrees of freedom and it solely depends on VAR specifications. The impulse response decomposition function is measured at a frequency given by the band  ℵ(d) = ∑ ω  ℵ( ω ). In this way, the desired frequency band for generalized decompositions of variance is estimated as: ( ∂ d)j,l=∑ ω  ρ j( ω )( f( ω ))j,l(12) Here ( f( ω ))j,l= δ−1 ll (( ℵ( ω )∑)jl )2/( ℵ( ω )∑ ℵ′( ω ))j.j is the generalized causation spectrum and  ρ j( ω ) = ( ℵ( ω )∑ ℵ′( ω ))j,j/(∅)j,j)is the weighted fraction and ∅=∑ ω ℵ( ω )∑ ℵ′( ω ). Thus, the measure of connectedness at a given desired frequency band is derived by substituting ( ∂ d)j,l into traditional measures. 3.4. Determinants of connectedness We used some generally focused financial and economic uncertainty indicators to determine green markets’ return and volatility connectedness with respect to the US stock market sectors and commodity 2 Note that in the empirical analysis, in addition to the standard GARCH(1,1)- model, the conditional volatility was modeled also based on an asymmetric representation of the data in the form of Glosten, Jagannathan, and Runkle (1993) GJR-GARCH-model, but this specification had no qualitative effect on our main results. M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 5 markets. The financial uncertainty indicators 3 used for the regression analysis are VIX, RUS, GVZ, EXG, EMR, and MOVE, whereas the economic uncertainty indicators 4 employed for the multivariate regressions are USEPU, UKEPU, USEQU, and INFD. The regression equation is given as follows: TCit =β0+β1∑Financialit +β2∑Economicit + ε it (13) where TC it denotes total return and volatility connectedness of market i at time t. β 0 is the intercept, whereas ε it is the error term. The component ∑Financial it represents the proxies of six financial variables employed in the study, while ∑Economic it reflects the four proxies of economic (and pandemic) uncertainty variables. 3.5. Data and descriptive statistics For investigating the return and volatility spillovers of green (financial) markets, US stock market sectors and commodity markets. For this purpose, we utilized the data from green markets in the form of S&P Green Bond Index (SPGB), Wilderhill Clean Energy Index (WHCL), World Renewable Energy Index (RENX), MSCI Global Green Building Index (MSGB), S&P Global Clean Energy Index (SPCL), MSCI ACWI Water Utility Index (MSWT), and EEX-EU CO 2 Emissions Index (EUCO). As representing the US stock market sectors we utilize the industry indexes from the healthcare (HLTH), consumer discretionary (CODC), energy (ENER), financials (FINL), industrials (INDS), communication services (COSV), materials (MATR), consumer staples (COST), information technology (TECH), and utilities (UTIL). Additionally, the commodity market indexes included in the study are from the crude oil (CWTI), heating oil (HTOL), natural gas (NTGS), gold (GOLD), silver (SLVR), copper (COPR), and wheat (WHET) spot markets. The daily data are taken from the Refinitiv/EIKON/Datastream, spanning observations form September 2010 to July 2021. Table 1 presents the descriptive statistics of return series (based on the natural log difference of the price/index series) where the highest average return from the green markets is reported by EUCO, followed by MSWT, RENX, MSGB, WHCL, and SPCL. Interestingly, SPGB, i.e., the S&P green bond index markets, showed zero average return for the sample period. TECH yields the highest mean return among US stock market sectors, followed by CODC and HLTH. Moderate mean returns are exhibited by INDS, FINL, MATR, COST, COSV, and UTIL. However, ENER showed negative average returns. Commodities like CWTI and GOLD revealed the highest mean returns, followed by COPR and SLVR. The lowest average returns are obtained from HTOL, NTGS, and WHET. Similar to the mean values, the variability of return series showed the highest value for the EUCO market, followed by RENX, WHCL, SPCL, and MSGB, whereas SPGB has the lowest variability. The US stock sectors with the highest variability in returns are ENER, FINL, MATR, INDS, and TECH. The other US sectors revealed moderate to low variability in the returns, such as CODC, COSV, UTIL, HLTH, and COST. Out of commodity market returns, the strongest variability is denoted in the cases of NTGS, CWTI, and HTOL, whereas the rest of the commodity markets showed moderate variability. Slightly negative skewness values indicate that green markets, US sectoral returns, and commodity markets have experienced substantial losses under unfavorable market conditions. The Jarque-Bera test reveals abnormal values in all series, pointing to the non-normal distribution of return series. Table 2 illustrates the descriptive statistics of the volatility series where EUCO yields the highest average volatility among the green asset markets, followed by WHCL and RENX. SPCL, MSWT, and MSGB experience moderate volatility, whereas SPGB marked has the lowest average volatility for the sample period. ENER sector, among US sectors, showed the highest average volatility, followed by FINL, MATR, TECH, INDS, and COSV. Conversely, CODC, UTIL, HLTH, and COST exhibited the lowest volatilities in average terms. NTGS and CWTI reported the highest mean volatilities among the commodity market assets, followed by SLVR, HTOL, WHET, and COPR, with GOLD indicating the lowest volatility. The volatility statistics reflect the highest variation in the RENX and EUCO green markets, followed by WHCL, SPCL, and MSGB. However, SPGB revealed the lowest variability in terms of average volatility. The US sectoral volatility series show comparable values confirming that these sectors are subject to volatile economic periods. Nevertheless, the commodity markets are denoted to experience the highest variability in the volatility for CWTI, while the remaining commodities have moderate to low variability in the volatilities. The positive values of (excess) skewness coefficients validate the existence of potential shocks, so the markets seem to have also been exposed to uncertainties. In addition, the Jarque-Bera test shows abnormally large values indicating that volatilities are not normally distributed. Fig. 1 presents the correlation heat maps between the green markets, US sectoral stocks, and commodities for both the return and volatility analyses, where the warm colour (orange) denotes the highest correlation whereas the cool colour (yellow) manifests low correlation. It is indicated in Fig. (1a) that WHCL and SPCL are highly correlated with the US stock market sectors, whereas slight correlation relationships are identified between the green markets and commodities. Correspondingly, US stock market sectors demonstrate high intercorrelations, whereas zero correlations are reported with commodities. Similarly, commodities also depict high correlations with the other commodities while no correlations are reported with green markets and US stock market sectors. Fig. (1b) presents the correlation heat-maps of volatility connectedness where some fragments of correlations are evident, reflecting within asset class high correlations and moderate to low correlations with other types of markets. Almost all green markets except SPGB, RENX, and EUCO reveal high correlations with the US stock market sectors and commodities. A larger fragment of high pairwise correlations among the US stock market sectors is presented, reiterating stronger correlations among similar classes of assets. Concurrently, commodities showcase sound correlations with other commodity classes and moderate to low correlations are reported with green markets and US sectors. 4. Empirical results 4.1. Return and volatility connectedness Fig. 2 illustrates more detailed results about the characteristics of the return connectedness of green markets with the US stock market sectors, and commodity markets, in terms of total connectedness (Fig. 2A) using the Diebold and Yilmaz (2012) spillover procedure, and short-run connectedness (Fig. 2B), and long-run connectedness measures (Fig. 2C) based on the Baruník and Kˇ rehlík (2018) approach. The total returns’ connectedness in Fig. 2A reveals pronounced intra-group connectedness among the US stock market sectors, whereas the green and commodity markets show low intra-group connectedness. The green markets consisting of WHCL, SPCL, and RENX indexes seem to form a distinct group within the same category market, while the other green markets show intra-group disconnection. Forming a separate cluster within the same category of green markets highlights the strong interconnectedness among the markets having similar features. For instance, WHCL, SPCL, and RENX share similar characteristics; therefore, their interconnectedness is significant. The high connectedness of green 3 VIX represents CBOE SPX Volatility Index, RUS denotes CBOE RUSSELL 2000 Volatility Index, GVZ is CBOE Gold Volatility Index, EXG is CBOE exchange index, EMR reflects CBOE Emerging Markets Volatility Index, and MOVE indicates ML MOVE 1 M Bond Volatility Index. All these implied volatilities are price indexes. 4 USEPU denotes US Economic Policy Uncertainty Index, UKEPU is UK Economic Policy Uncertainty Index, USEQU is US Equity related Economic Uncertainty, and INFD is the Infectious Disease EMV Tracker. All variables are economic series. M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 6 markets corroborates with the results of Naeem et al. (Naeem, Adekoya, & Oliyide, 2021; Naeem, Rabbani, et al., 2021), who also reported strong interconnection among the green market segments. However, our findings are against Elsayed, Nasreen, and Tiwari (2020), who reported similar results where green markets are lowly connected with various markets. However, the strong disconnection of remaining green markets highlights their diversification potential for several risky investments consistent with with Reboredo et al. (2020) and Arif, Hasan, et al. (2021), who also report strong diversification benefits of green markets. The US stock market sectors show strong intra-market connectedness demonstrating higher dependence of US sectors with frequent bidirectional spillovers. INDS is receiving spillovers from ENER, FINL, COSV, COOC, and HLTH, indicating in general that INDS is an aggressive industry that receives surmounted spillovers from others (see also Ngene, 2021). The commodity markets form two clusters where the precious metals are strongly interconnected, and heating and crude oil form another cluster among the commodity markets, whereas the rest of the commodities are disconnected from the network. The clustering of commodities aligns with the studies of Caporin et al. (2021) and Balli et al. (2019), where similar commodities were found to be clustered into distinct groups due to their comparable features. Notably, there is weak inter-group connectedness between the green markets, US stock market sectors, and commodities, where UTIL and MSWT experience bidirectional spillovers and concur the findings of Diebold, Liu, and Yilmaz (2017). This contends that inherent parallel characteristics of markets are more connected as compared to those with Table 1 Preliminary statistics for the return series. Group Symbol Mean Maximum Minimum Std. Dev. Skewness Kurtosis Jarque-Bera Green Markets SPGB 0.000 2.557 −3.091 0.390 −0.354 9.219 4645.526*** WHCL 0.024 13.399 −16.239 1.951 −0.513 9.678 5412.619*** SPCL 0.013 11.035 −12.498 1.459 −0.615 11.361 8468.35*** RENX 0.035 42.217 −41.627 2.302 −0.519 129.948 1911188*** MSGB 0.031 9.089 −11.740 1.029 −1.453 24.441 55,515.12*** MSWT 0.036 9.750 −9.411 1.030 −0.363 15.581 18,831.460*** EUCO 0.044 21.060 −44.655 3.153 −1.004 21.194 39,729.96*** US Sectors HLTH 0.055 7.314 −10.528 1.034 −0.455 13.213 12,467.71*** CODC 0.064 8.286 −12.877 1.134 −0.978 15.944 20,320.99*** ENER −0.001 15.111 −22.417 1.715 −0.978 23.975 52,622.52*** FINL 0.042 12.425 −15.071 1.429 −0.612 18.273 27,840.03*** INDS 0.045 12.001 −12.155 1.226 −0.676 17.069 23,687.51*** COSV 0.032 8.802 −11.030 1.111 −0.540 12.528 10,904.09*** MATR 0.036 11.003 −12.147 1.303 −0.603 12.842 11,659.36*** COST 0.035 8.075 −9.690 0.861 −0.427 20.435 36,132.09*** TECH 0.074 11.300 −14.983 1.297 −0.613 16.989 23,385.67*** UTIL 0.027 12.320 −12.265 1.096 −0.329 25.549 60,345.49*** Commodities CWTI 0.015 22.394 −28.221 2.573 −0.307 28.210 75,407.24*** HTOL 0.003 18.196 −19.996 2.011 −0.589 17.197 24,066.48*** NTGS 0.002 26.749 −18.055 2.920 0.416 8.684 3913.353*** GOLD 0.013 5.775 −9.821 1.015 −0.653 10.250 6435.749*** SLVR 0.009 8.948 −19.518 1.937 −0.998 11.392 8823.807*** COPR 0.010 6.810 −7.591 1.336 −0.144 5.452 723.0313*** WHET 0.002 12.929 −9.223 1.830 0.423 5.817 1026.071*** Note: *** indicates significance at 1%. Table 2 Preliminary statistics for volatility series. Group Symbol Mean Maximum Minimum Std. Dev. Skewness Kurtosis Jarque-Bera Green Markets SPGB 0.356 1.092 0.168 0.154 1.803 6.891 3337.819*** WHCL 1.775 7.731 0.947 0.792 2.837 15.153 21,330.25*** SPCL 1.302 6.706 0.613 0.643 3.271 19.669 38,025.42*** RENX 1.764 31.417 1.042 1.508 10.497 148.007 2545725*** MSGB 0.867 5.533 0.432 0.515 4.089 26.796 75,075.67*** MSWT 0.912 5.244 0.602 0.385 6.910 65.360 483,797.6*** EUCO 2.941 15.890 1.143 1.435 2.605 15.951 23,109.54*** US Sectors HLTH 0.922 5.224 0.509 0.433 4.405 32.411 111,779.1*** CODC 1.000 6.776 0.528 0.514 4.178 31.678 105,804.1*** ENER 1.474 8.943 0.651 0.878 3.998 26.162 71,200.36*** FINL 1.216 8.930 0.639 0.692 4.643 34.956 131,316.3*** INDS 1.065 6.924 0.557 0.575 4.707 35.858 138,536.6*** COSV 1.024 5.537 0.694 0.356 5.300 46.790 240,716.2*** MATR 1.166 6.463 0.598 0.570 4.012 27.271 77,492.92*** COST 0.747 5.626 0.437 0.379 6.450 63.298 450,889.5*** TECH 1.144 8.126 0.576 0.591 4.513 36.567 143,268.4*** UTIL 0.927 6.360 0.540 0.510 6.524 56.460 359,092.9*** Commodities CWTI 2.184 14.036 1.000 1.359 4.503 29.119 90,517.12*** HTOL 1.804 8.587 0.919 0.864 3.348 18.070 32,244.93*** NTGS 2.779 9.042 1.575 0.895 1.850 8.332 4993.389*** GOLD 0.976 2.061 0.617 0.262 1.454 5.221 1586.948*** SLVR 1.828 5.019 1.032 0.645 1.391 5.121 1450.875*** COPR 1.314 2.987 0.854 0.267 2.051 10.581 8811.059*** WHET 1.794 3.198 1.089 0.364 0.765 3.485 305.1684*** Note: *** indicates significance at 1%. M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 7 Fig. 1. Correlation heat-maps. Note: This figure shows the correlation heat-maps among Green Markets, US Sectors and Commodities. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 8 dissimilar features. Gold has a unidirectional spillover with SPGB, whereas CWTI and HTOL experience unidirectional connectedness with ENER. A closer look at Fig. 2B manifesting the short-run connectedness of green markets, US sectors, and commodities, following the Baruník and Kˇ rehlík (2018) approach, reveals a parallel connectedness pattern as depicted by the total connectedness. However, the connectedness pattern in the long-run (Fig. 2C) shows strong inter-group connectedness and weak intra-group spillovers. The strong inter-group connectedness among markets concurs with the results of Zhao et al. (2021), documenting strong spillovers in the long-run. Moreover, spillovers are significant for green markets highlighting their net transmitting role in influencing the network spillovers. These findings align with the results of Arif et al. (Arif, Hasan, et al., 2021; Arif, Naeem, et al., 2021) and Naeem, Nguyen, Nepal, Ngo, and Taghizadeh-Hesary (2021), where the green markets exhibit strong spillovers with respect to other markets. Fig. 3 illustrates the volatility connectedness of green markets, US stock market sectors, and commodities, where the system-wide connectedness indicates high inter-group spillovers transmitting from green markets to US sectors and moderate spillovers from commodities to the stock market sectors. Hence, the green and commodity markets are net transmitters in the total connectedness, whereas the US stock market sectors are net recipients of volatility spillovers. Meanwhile, the intra-group volatility connectedness is significant in the US sectors, and modest to low intra-group volatilities are observed in the green markets and commodities. Interestingly, the volatility connectedness in the short-run (Fig. 3B) is prominent in the US stock market sectors, whereas mild spillovers are evident in the green markets and commodities. In our data, the volatility connectedness in the long-run (Fig. 3C) exhibits similar spillover patterns as shown in Fig. 3A, where the green markets and commodities are transmitting spillovers to various US stock market sectors, which indicates their outperformance during volatile times. In contrast, strong intra-group volatility connectedness of the US stock market sectors and the net recipient characteristics of volatility spillovers highlight their extreme exposure to the uncertainties of the economic environment. In line with Mensi et al. (2021), our findings confirm that the US sectoral returns exhibit intense spillovers possibly Fig. 2. Return connectedness network among green asset markets, US stock market sectors and commodity markets. Note: This Figure shows the return connectedness among green asset markets, US stock market sectors and commodity markets. Total connectedness network is estimated using the Diebold and Yilmaz (2012) procedure, whereas short-run and long-run connectedness networks are estimated using the Barunik and Krehlik (2018) approach. Each group is represented by a colour. We only report the values larger than the average of the 100 largest individual pairwise connectedness measures. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 15 volatilities of exchange rates. EMR drives negatively the connectedness of the markets in the long-run, implying that the connectedness becomes weaker when the EMR upsurges. In the case of MOVE, the increase in the bond market volatility reduces the volatility connectedness of markets in view of the total and long-run perspectives. Concerning the general economic uncertainty effects, only UKEPU and INFD show up as determinants of the total volatility connectedness, suggesting an increase in the UKEPU and INFD enhances the volatility connectedness of the markets. Given this, we report that the financial indicators drive substantially, whereas the economic indicators drive more modestly the connectedness between green asset markets, US stock market sectors, and commodity markets. 5. Conclusions and policy implications We aimed to investigate the time-frequency return and volatility connectedness of green markets, US sectors, and commodity markets using daily data from September 2010 to July 2021. We utilize the GARCH (1,1) model to estimate the conditional volatility of the sample of all the analyzed return series. Using the Diebold and Yilmaz (2012) and Baruník and Kˇ rehlík (2018) methods, we analyzed the timefrequency connectedness of the markets given their return and volatility connections. Furthermore, we analyzed the role of some relevant financial and economic uncertainty indicators prominently underlining the connectedness. Our study embraced support from modern portfolio theory for diversifying the risk of financial markets. On the other hand, the market reaction hypothesis posits that various financial and economic uncertainties drive the underlying spillover network of markets echoing the heterogenous response of investors toward these factors. Our results indicate complex intra-group return connectedness and mild inter-group return connectedness. The US stock market sectors revealed strong intra-group spillovers for the part of both the time and frequency return connectedness. On the other hand, the finding of system-wide volatility connectedness regarding the total, short- and long-term relationships indicates a high transmission of spillovers from the green asset markets and commodity markets to the US stock market sectors and strong intra-group connectedness among the US sectors. Based on our results, we propose the following implications for the policymakers and investors or portfolio managers regarding their asset allocation, risk mitigation, and other financial market actions. First of all, the time-varying attributes of return and volatility connectedness exhibit that the analyzed markets have suffered from strong shocks during unexpected economic periods such as the European Debt Crisis, Shale oil crisis, Chinese stock market crash, and COVID-19 pandemic. Spillovers between the markets intensify when the markets experience stressful times, whereas the troughs in the connectedness graphs are in connection to the stable market conditions. Moreover, the short-run return spillovers seem to dominate the long-run spillovers reflecting the emergence of financial contagion during economically most fragile times. However, the time-varying results of volatility connectedness stress our finding that higher volatility is associated with long-term developments, with several significant events affecting the market spillovers. Hence, the long-run volatility spillovers dominate the shortrun spillovers based on our results. Furthermore, time-varying NET return and volatility connectedness results characterize the US stock market sectors as the NET transmitters of spillovers. In contrast, the green asset markets and commodity markets can be denoted as NET receivers of return and volatility spillovers with significant time variations. While assessing the driving forces of financial and economic uncertainty indicators underscoring the connectedness, we found a Table 3 Impact of financial and economic uncertainty indicators on time and frequency return connectedness. Indicator Total Short Long C 87.812*** 72.458*** 15.348*** (1.759) (1.499) (0.625) Financial VIX −0.173* −0.402*** 0.229*** (0.095) (0.084) (0.042) RUS 0.298*** 0.422*** −0.124*** (0.091) (0.078) (0.034) GVZ 0.414*** 0.450*** −0.035* (0.058) (0.055) (0.020) EXG −0.199*** −0.173*** −0.025*** (0.009) (0.007) (0.003) EMR −0.057 −0.051 −0.007 (0.060) (0.054) (0.022) MOVE −0.039*** −0.064*** 0.025*** (0.014) (0.014) (0.006) Economic USEPU 0.002 0.004* −0.002** (0.002) (0.002) (0.001) UKEPU 0.002** 0.001 0.001*** (0.001) (0.001) (0.000) USEQU −0.002 −0.003** 0.001 (0.002) (0.001) (0.001) INFD 0.180*** 0.201*** −0.021 (0.046) (0.043) (0.013) R 2 0.747 0.778 0.344 Adjusted R 2 0.746 0.777 0.342 Note: This regression is based on HAC (Newey-West) heteroscedasticityconsistent standard errors & covariance. The table presents the results for the role of financial and economic uncertainty indicators, based on using log changes of the indexes for the US stock market volatility (VIX), UK stock market volatility (RUS), Gold market volatility (GVZ), CBOE Exchange Index (EXG), Emerging markets volatility (EMR), Treasury market volatility (MOVE), US Economic Policy Uncertainty (USEPU), UK Economic Policy Uncertainty (UKEPU), US Equity related Uncertainty (USEQU), and the Infectious Diseases Tracker (INFD), respectively. The values in () are standard errors. C refers to the constant term in the regression equation. The asterisks *, ** and *** stand for the 10, 5 and 1% risk levels of significance, respectively. Table 4 Impact of financial and economic uncertainty indicators on time and frequency volatility connectedness. Indicator Total Short Long C 73.452*** 11.431*** 67.103*** (2.202) (1.201) (3.204) Financial VIX −0.358*** −0.068 −0.017 (0.126) (0.061) (0.162) RUS 0.479*** −0.170** 0.597*** (0.116) (0.071) (0.165) GVZ 0.640*** −0.004 0.558*** (0.075) (0.035) (0.086) EXG −0.087*** −0.025*** −0.038** (0.011) (0.007) (0.018) EMR −0.082 0.078** −0.248*** (0.079) (0.037) (0.096) MOVE −0.083*** 0.006 −0.048** (0.019) (0.010) (0.021) Economic USEPU 0.002 −0.002 −0.001 (0.003) (0.002) (0.004) UKEPU 0.004*** 0.000 0.001 (0.001) (0.000) (0.001) USEQU 0.001 0.003** −0.005* (0.002) (0.002) (0.003) INFD 0.202*** 0.057 0.043 (0.051) (0.041) (0.075) R 2 0.637 0.237 0.550 Adjusted R 2 0.635 0.234 0.549 Note: This regression is based on HAC (Newey-West) heteroscedasticityconsistent standard errors & covariance. The table presents the results for the role of financial and economic uncertainty indicators, based on using log changes of the indexes for the US stock market volatility (VIX), UK stock market volatility (RUS), Gold market volatility (GVZ), CBOE Exchange Index (EXG), Emerging markets volatility (EMR), Treasury market volatility (MOVE), US Economic Policy Uncertainty (USEPU), UK Economic Policy Uncertainty (UKEPU), US Equity related Uncertainty (USEQU), and the Infectious Diseases Tracker (INFD), respectively. The values in () are standard errors. C refers to the constant term in the regression equation. The asterisks *, ** and *** stand for the 10, 5 and 1% risk levels of significance, respectively. M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 16 positive driving influence of Russell Index and Gold, but a negative effect of currency and bond markets on the return and volatility connectedness of the markets focused. We report that the UK economic policy uncertainty and infectious disease tracker positively affect the return and volatility connectedness of markets for the part of general economic uncertainty factors. These findings are crucial for policymakers for developing and assessing their future policies, particularly when investors have serious concerns about economic and financial stability in the face of an economic downturn. Moreover, a clear picture is provided to the policymakers through a nuanced approach of time and frequency connectedness for the return and volatilities, independently. The findings can set a benchmark for the policymakers in restructuring and revisiting their outdated policies about the stability of the financial and commodity markets. For investors, the results are vital for risk mitigation and streamlining their portfolios as this study would also help intuitive decision-making when predicting the future returns and trying to offset the portfolio risks by adding diversifiers in their portfolios. For example, for the green asset market investors, the findings are appealing because they give a clearer picture of the connections between the green asset markets and other relevant financial market sectors, both in terms of return and volatility connections. Finally, the last step of our empirical analysis revealed that the spillovers seem to shift based on market ups and downs during unexpected financial and economic stress conditions. Hence, investors have to clearly distinguish between the short- and long-run spillovers, both with respect to the return and volatility connectedness. This should also help the policymakers in revealing the risk-adjusted potential of the small fish (green markets) in mitigating the risks of other big ponds (US stock market sectors and commodities) markets. Considering the determinants of return and volatility connectedness in various financial and economic uncertainty indicators, our findings can potentially benefit investors and policymakers for risk and return predictability during unfavorable financial markets and aggregate economic circumstances. Investors and financial market participants can consider these driving forces for their portfolio management, risk management, and asset allocation decisions. In addition, our timevarying analysis revealed that it is essential for the traders and portfolio managers to make adjustments in their existing investment positions during varying market conditions. Our detailed analysis of the spillover structure provides significant insights to the macro-prudential regulators to protect the most fragile markets and select the appropriate policies and regulatory attempts to preserve the interests of investors during unexpected financial and economic conditions. CRediT authorship contribution statement Muhammad Abubakr Naeem: Conceptualization, Methodology, Software, Formal analysis, Visualization, Writing – review & editing. Sitara Karim: Conceptualization, Writing – original draft, Writing – review & editing. Gazi Salah Uddin: Conceptualization, Data curation, Writing – review & editing, Project administration. Juha Junttila: Writing – original draft, Writing – review & editing, Supervision, Funding acquisition. Acknowledgement Earlier versions of this paper were presented at the Link¨ oping University, Sweden; University of Jyv¨ askyl¨ a School of Business and Economics, Jyv¨ askyl¨ a, Finland and as an invited keynote talk at the International Conference on Future Outlook of Oil market and Fiscal Stability, Nov 17-18, 2021, Ural Federal University. The third co-author is thankful for the academic research support provided by the Asian Development Bank and visiting fellow grant provided by University of Jyv¨ askyl¨ a School of Business and Economics, Jyv¨ askyl¨ a, Finland. Authors are thankful to the University of Jyv¨ askyl¨ a School of Business and Economics for the academic facilities provided to Third co-author during his stay at Jyv¨ askyl¨ a where important parts of this research work were completed. We are also grateful to the Editor Brian Lucey and two anonymous referees for their helpful comments that greatly improved the quality of this paper. References Abakah, E. J. A., Addo, E., Gil-Alana, L., & Tiwari, A. K. (2021). Re-examination of international bond market dependence: Evidence from a pair copule approach. International Review of Financial Analysis, 74, Article 101678. Abbas, G., Hammoudeh, S., Shahzad, S. J. H., Wang, S., & Wei, Y. (2019). Return and volatility connectedness between stock markets and macroeconomic factors in the G- 7 countries. Journal of Systems Science and Systems Engineering, 28(1), 1–36. Ahmad, W., Hernandez, J. A., Saini, S., & Mishra, R. K. (2021). The US equity sectors, implied volatilities, and COVID-19: What does the spillover analysis reveal? Resources Policy, 72, Article 102102. Andersen, T. M., Bhattacharya, J., & Liu, P. (2020). Resolving intergenerational conflict over the environment under the Pareto criterion. Journal of Environmental Economics and Management, 100, Article 102290. Apostolakis, G. N., Floros, C., Gkillas, K., & Wohar, M. (2021). Political uncertainty, COVID-19 pandemic and stock market volatility transmission. Journal of International Financial Markets, Institutions and Money, 74, Article 101383. Arif, M., Hasan, M., Alawi, S. M., & Naeem, M. A. (2021). COVID-19 and time-frequency connectedness between green and conventional financial markets. Global Finance Journal, 49, Article 100650. Arif, M., Naeem, M. A., Farid, S., Nepal, R., & Jamasb, T. (2021). Diversifier or more? Hedge and safe haven properties of green bonds during COVID-19. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3782126. Bahloul, S., & Khemakhem, I. (2021). Dynamic return and volatility connectedness between commodities and Islamic stock market indices. Resources Policy, 71, Article 101993. Bakas, D., & Triantafyllou, A. (2018). The impact of uncertainty shocks on the volatility of commodity prices. Journal of International Money and Finance, 87, 96–111. Balli, F., Hasan, M., Ozer-Balli, H., & Gregory-Allen, R. (2021). Why do U.S. uncertainties drive stock market spillovers? International evidence. International Review of Economics and Finance, 76, 288–301. Balli, F., Naeem, M. A., Shahzad, S. J. H., & de Bruin, A. (2019). Spillover network of commodity uncertainties. Energy Economics, 81, 914–927. Baruník, J., & Kˇ rehlík, T. (2018). Measuring the frequency dynamics of financial connectedness and systemic risk. Journal of Financial Econometrics, 16(2), 271–296. Blundell-Wignall, A. (2012). Solving the financial and sovereign debt crisis in Europe. OECD Journal: Financial Market Trends, 2011(2), 201–224. Bouri, E., Cepni, O., Gabauer, D., & Gupta, R. (2021). Return connectedness across asset classes around the COVID-19 outbreak. International Review of Financial Analysis, 73, Article 101646. Caporin, M., Naeem, M. A., Arif, M., Hasan, M., Vo, X. V., & Shahzad, S. J. H. (2021). Asymmetric and time-frequency spillovers among commodities using high-frequency data. Resources Policy, 70, Article 101958. CBI. (2020). Green bonds market summary H1–2020. Climate Bonds Initiative in Association with HSBC Climate Change Centre of Excellence. Daubanes, J. X., Shema, F. M., & Rochet, J.-C. (2022). Why Do Firms Issue Green Bonds, MIT Center for Energy and Environmental Policy Research (CEEPR) Working Paper 2022–001, January 2022. Diebold, F. X., Liu, L., & Yilmaz, K. (2017). Commodity connectedness. National Bureau of economic research working paper no 23685. Diebold, F. X., & Yilmaz, K. (2012). Better to give than to receive: Predictive directional measurement of volatility spillovers. International Journal of Forecasting, 28(1), 57–66. Diebold, F. X., & Yılmaz, K. (2014). On the network topology of variance decompositions: Measuring the connectedness of financial firms. Journal of Econometrics, 182(1), 119–134. Elsayed, A. H., Naifar, N., Nasreen, S., & Tiwari, A. K. (2022). Dependence structure and dynamic connectedness between green bonds and financial markets: Fresh insights from time-frequency analysis before and during COVID-19 pandemic. Energy Economics, 105842. Elsayed, A. H., Nasreen, S., & Tiwari, A. K. (2020). Time-varying comovements between energy market and global financial markets: Implication for portfolio diversification and hedging strategies. Energy Economics, 90, Article 104847. https://doi.org/ 10.1016/j.eneco.2020.104847 Farid, S., Kayani, G. M., Naeem, M. A., & Shahzad, S. J. (2021). Intraday volatility transmission among precious metals, energy, and stocks during the COVID-19 pandemic. Resources Policy, 72, Article 102101. Ferrer, R., Shahzad, S. J. H., & Soriano, P. (2021). Are green bonds a different asset class? Evidence from time-frequency connectedness analysis. Journal of Cleaner Production, 125988. Glosten, L. R., Jagannathan, R., & Runkle, D. E. (1993). On the relation between the expected value and the volatility of the nominal excess returns on stocks. The Journal of Finance, 48(5), 1779–1801. Kang, S., Hernandez, J. A., Sadorsky, P., & McIver, R. (2021). Frequency spillovers, connectedness, and the hedging effectiveness of oil and gold for US sector ETFs. Energy Economics, 99, Article 105278. M.A. Naeem et al. International Review of Financial Analysis 83 (2022) 102283 17 Karim, S., Khan, S., Mirza, N., Alawi, S. M., & Taghizadeh-Hesary, F. (2022). Climate finance in the wake of COVID-19: Connectedness of clean energy with conventional energy and regional stock markets. Climate Change Economics, 2240008. Karim, S., Lucey, B. M., & Naeem, M. A. (2022). The dark side of bitcoin: Do emerging Asian Islamic markets subdue the ethical risk? (Available at SSRN 4025831). Karim, S., Lucey, B. M., Naeem, M. A., & Uddin, G. S. (2022). Examining the interrelatedness of NFTs, DeFi tokens and cryptocurrencies. Finance Research Letters, 102696. Karim, S., & Naeem, M. A. (2021). Clean energy, Australian electricity markets, and information transmission. Energy Research Letters, 3, 29973. Early View. Karim, S., & Naeem, M. A. (2022). Do global factors drive the interconnectedness among green, Islamic and conventional financial markets? International Journal of Managerial Finance. https://www.emerald.com/insight/1743-9132.htm. Khalfaoui, R., Jabeur, S. B., & Dogan, B. (2022). The spillover effects and connectedness among green commodities, bitcoins, and US stock markets: Evidence from the quantile VAR network. Journal of Environmental Management, 306, Article 114493. Kilian, L., & Zhou, X. (2018). Modeling fluctuations in the global demand for commodities. Journal of International Money and Finance, 88, 54–78. Leitao, J., Ferreira, J., & Santibanez-Gonzalez, E. (2021). Green bonds, sustainable development and environmental policy in the European Union carbon market. Business Strategy and the Environment, 1–14. Markowitz, H. (1952). Portfolio selection, the. Journal of Finance, 7/(1), 77–91. Marshall, B. R., Nguyen, H. T., Nguyen, N. H., Visaltanachotin, N., & Young, M. (2021). Do climate risks matter for green investment? Journal of International Financial Markets, Institutions and Money, 101438. Mensi, W., Naeem, M. A., Vo, X. V., & Kang, S. H. (2022). Dynamic and frequency spillovers between green bonds, oil and G7 stock markets: Implications for risk management. Economic Analysis and Policy, 73, 331–344. Mensi, W., Nekhili, R., Vo, X. V., Suleman, T., & Kang, S. H. (2021). Asymmetric volatility connectedness among U.S. stock sectors. North American Journal of Economics and Finance, 56, Article 101327. Mensi, W., Rehman, M. U., & Vo, X. V. (2020). Spillovers and comovements between precious metals and energy markets: Implications on portfolio management. Resources Policy, 69, Article 101836. Naeem, M. A., Adekoya, O. B., & Oliyide, J. A. (2021). Asymmetric spillovers between green bonds and commodities. Journal of Cleaner Production, 128100. Naeem, M. A., Farid, S., Qureshi, F., & Taghizadeh-Hesary, F. (2022). Global factors and the transmission between United States and emerging stock markets. International Journal of Finance & Economics.. https://doi.org/10.1002/ijfe.2604 Naeem, M. A., Hasan, M., Arif, M., Balli, F., & Shahzad, S. J. H. (2020). Time and frequency domain quantile coherence of emerging stock markets with gold and oil prices. Physica A: Statistical Mechanics and its Applications, 553, Article 124235. Naeem, M. A., & Karim, S. (2021). Tail dependence between bitcoin and green financial assets. Economics Letters, 110068. Naeem, M. A., Karim, S., Jamasb, T., & Nepal, R. (2022). Risk transmission between green markets and commodities (Available at SSRN). Naeem, M. A., Nguyen, T. T. H., Nepal, R., Ngo, Q. T., & Taghizadeh-Hesary, F. (2021). Asymmetric relationship between green bonds and commodities: Evidence from extreme quantile approach. Finance Research Letters, 101983. Naeem, M. A., Pham, L., Senthilkumar, A., & Karim, S. (2022). Oil shocks and BRIC markets: Evidence from extreme quantile approach. Energy Economics, 105932. Naeem, M. A., Rabbani, M. R., Karim, S., & Billah, S. M. (2021). Religion vs ethics: Hedge and safe haven properties of Sukuk and green bonds for stock markets pre-and during COVID-19. International Journal of Islamic and Middle Eastern Finance and Management.. https://doi.org/10.1108/IMEFM-06-2021-0252. In press. Ngene, G. M. (2021). What drives dynamic connectedness of the U.S equity sectors during different business cycles. North American Journal of Economics and Finance, 58, Article 101493. Nguyen, T. T. H., Naeem, M. A., Balli, F., Balli, H. O., & Vo, X. V. (2020). Time-frequency comovement among green bonds, stocks, commodities, clean energy, and conventional bonds. Finance Research Letters, 101739. Omisore, I., Yusuf, M., & Christopher, N. (2012). The modern portfolio theory as an investment decision tool. Journal of Accounting and Taxation, 4(2), 19–28. Pesaran, H. H., & Shin, Y. (1998). Generalized impulse response analysis in linear multivariate models. Economics Letters, 58(1), 17–29. Pham, L., & Huynh, T. L. D. (2020). How does investor attention influence the green bond market? Finance Research Letters, 35, Article 101533. Prokopczuk, M., Stancu, A., & Symeonidis, L. (2019). The economic drivers of commodity market volatility. Journal of International Money and Finance, 98, Article 102063. Reboredo, J. C., Ugolini, A., & Aiube, F. A. L. (2020). Network connectedness of green bonds and asset classes. Energy Economics, 86, Article 104629. Rufino, C. C. (2018). Long-run linkages of ASEAN+3 floating currencies. Business & Economics Review, 27, 1–14. Russo, A., Mariani, M., & Caragnano, A. (2021). Exploring the determinants of green bond issuance: Going beyond the long-lasting debate on performance consequences. Business Strategy and the Environment, 30(1), 38–59. Saeed, T., Bouri, E., & Alsulami, H. (2021). Extreme return connectedness and its determinants between clean/green and dirty energy investments. Energy Economics, 96, Article 105017. Salisu, A. A., Raheem, I. D., & Vo, X. V. (2021). Assessing the safe haven property of the gold market during COVID-19 pandemic. International Review of Financial Analysis, 74, Article 101666. Shahzad, S. J. H., Bouri, E., Kristoufek, L., & Saeed, T. (2021). Impact of the COVID-19 outbreak on the US equity sectors: Evidence from quantile return spillovers. Financial Innovation, 7(14), 1–23. Tang, D. Y., & Zhang, Y. (2020). Do shareholders benefit from green bonds? Journal of Corporate Finance, 61, Article 101427. Tiwari, A. K., Abakah, E. J. A., Gabauer, D., & Dwumfour, R. A. (2022). Dynamic spillover effects among green bond, renewable energy stocks and carbon markets during COVID-19 pandemic: Implications for hedging and investments strategies. Global Finance Journal, 51, Article 100692. Umar, Z., Adekoya, O. B., Oliyide, J. A., & Gubareva, M. (2021). Media sentiment and short stocks performance during a systemic crisis. International Review of Financial Analysis, 78(2021), Article 101896. Urom, C., Mzoughi, H., Abid, I., & Brahim, M. (2021). Green markets integration in different time scales: A regional analysis. Energy Economics, 98, Article 105254. Wang, G. J., Xie, C., Zhao, L., & Jiang, Z. Q. (2018). Volatility connectedness in the Chinese banking system: Do state-owned commercial banks contribute more? Journal of International Financial Markets, Institutions and Money, 57, 205–230. Wang, J., Chen, X., Li, X., Yu, J., & Zhong, R. (2020). The market reaction to green bond issuance: Evidence from China. Pacific-Basin Finance Journal, 60, Article 101294. Womack, B. (2017). International crises and China’s rise: Comparing the 2008 global financial crisis and the 2017 global political crisis. The Chinese Journal of International Politics, 10(4), 383–401. Zhao, Y., Umar, Z., & Vo, X. V. (2021). Return and volatility connectedness of Chinese onshore, offshore, and forward exchange rate. Journal of Futures Markets, 41(11), 1843–1860. M.A. Naeem et al.