Next Article in Journal
Advancing Hydrogen: A Closer Look at Implementation Factors, Current Status and Future Potential
Previous Article in Journal
Changes in Commercial Dendromass Properties Depending on Type and Acquisition Time
Previous Article in Special Issue
Nuclear Energy and Financial Development for a Clean Environment: Examining the N-Shaped Environmental Kuznets Curve Hypothesis in Top Nuclear Energy-Consuming Countries
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Nexus between Green Bonds and European Banks: A Cross-Quantilogram Approach

1
“Victor Slavescu” Centre for Financial and Monetary Research, Romanian Academy, 050711 Bucharest, Romania
2
Department of International Business and Economics, Bucharest University of Economic Studies, 010404 Bucharest, Romania
3
Institute for Economic Forecasting, Romanian Academy, 050711 Bucharest, Romania
*
Author to whom correspondence should be addressed.
Energies 2023, 16(24), 7974; https://doi.org/10.3390/en16247974
Submission received: 5 October 2023 / Revised: 21 November 2023 / Accepted: 6 December 2023 / Published: 8 December 2023
(This article belongs to the Special Issue Financial Development and Energy Consumption Nexus II)

Abstract

:
Financial markets have the potential to magnify the adverse impacts of carbon-intensive assets, mainly in the case of a swift and unforeseen shift toward a low-carbon economy. Given that green bonds are already in the process of standardization and actively support the funding of environmental goals, this paper aims to explore their relationship with the European banking system. To achieve this objective, we utilize a cross-quantilogram approach, analyzing daily data gathered from July 2014 to January 2021 and examining bi-directional dependence. Our unique contribution lies in revealing the relationships between the green bond index and the stock market dynamics of European banks compared to their relationships with conventional stock market indices, which is a novel endeavor to the best of our knowledge. The results are consistent with prior research findings regarding the relationships between the green bond index and various companies and financial assets. These results confirm that other financial instruments impact green bonds, whereas the influence exerted by green bonds on other assets is minimal. Additionally, our study provides evidence indicating that the COVID-19 pandemic has altered the connections between these financial assets.

1. Introduction

Financial markets can enhance the adverse effects of carbon-intensive assets, which occurs primarily if the transition to a low-carbon economy occurs quickly and unexpectedly. The contagion can influence the corporate bond market if investors cannot sell energy-related debt due to a lack of liquidity. Stock price corrections could increase market volatility, leading to the stop loss option’s exercise, helping amplify already declining price movements.
The economic literature explores three distinct forms of potential contagion or correlation within financial markets [1,2,3]. Bottom-up contagion can emanate from the company’s losses and blocked assets. Among the factors that cause such a contagion may be physical, policy, and regulatory changes. Capital flight can occur due to the degradation of natural capital in a particular country or region, or it may occur because of the possibility of a future increase in the value of natural capital stocks in different geographical areas. In general, it occurs as a result of the transition from non-renewable to renewable resources. Such capital movements could harm the overall macroeconomic environment, affecting inflation, exchange rates, or fiscal policies, especially in countries with low domestic investment levels and deficits. Global trade flows and supply chains, along with the degradation of natural capital, can impair financial stability through hazard globalization. This finding is underlined by Sternberg [4], who shows that this situation can lead to a deterioration of social, economic, and political aspects, ultimately influencing overall financial stability. The author points out that a drought in the wheat-growing region of China severely affected the markets of Egypt, the world’s largest importer of wheat. This increase in food prices has become one of the triggers for civil protests. Moreover, it was documented in the economic literature that there are spillovers between the energy sector and other economic sectors [5].
Conversely, when the advancement of the green bond market loses momentum [6], the European countries exhibit differences in their following of climate and energy objectives [7], and a deeper analysis regarding the green bond market is called upon in the literature [8].
These studies present a solid argument for researching the many aspects accompanying the development of a relatively new market, green financing. Given that green bonds are currently in the process of standardization and play an active role in financing environmental goals, coupled with their traceability features, our objective is to explore their relationship with the financial sector. The intersection of green bonds and European banks represents a pivotal area of inquiry, offering insights into the dynamic relationship between sustainable financial instruments and the banking sector within the European context. In exploring this nexus, our study employs a cross-quantilogram approach, a methodological framework that has been designed to assess the strengths of connections and dependencies, particularly in terms of tail distributions. This approach enhances our understanding of the intricate linkages between green bonds and European banks and provides a nuanced perspective on how these relationships evolve under different market conditions. Green bonds introduce a unique risk–return profile; understanding how financial markets engage with these instruments involves assessing implications for risk assessment and management within the context of environmentally conscious investments. By delving into this unique crossroads, our research contributes to the broader discourse on sustainable finance, shedding light on the implications and potential impact of green bonds within the framework of European banking institutions.
Our study significantly enhances the existing literature through several notable contributions. Firstly, we introduce a robust methodology for evaluating the strengths of financial relationships, with a particular emphasis on distinguishing them from associations with conventional stock market indices. We employ a sophisticated cross-quantilogram methodology, building upon previous advancements [9]. Secondly, our research pioneers a bidirectional analysis that focuses on exploring connections at the tail distribution level. This innovative/subtle approach allows us to unveil non-linear dependencies, particularly amid extreme market conditions. By delving into the intricacies of these relationships, we provide a nuanced understanding that goes beyond traditional linear analyses. Thirdly, in response to the unprecedented challenges posed by COVID-19, we contribute to the literature by devising a methodological framework to investigate the impact of the pandemic on these connections. Our framework enables us to offer compelling evidence regarding the changes in these relationships, shedding light on the dynamic nature of financial interactions during times of crisis. Overall, our study expands the methodological toolkit for relationship evaluation and deepens our insights into the complexities of financial connections, especially in the context of evolving market conditions.
Our analysis features the following structure. First, we summarize the primary literature and the corresponding results, to identify the research gap and position our approach. The data and research methodology are described afterward. We then present our results and explain their importance in light of the findings from other studies. We conclude with a summary of the primary outcomes, the limitations of our study, some proposals for future research, and practice recommendations.

2. Literature Review

Studying the links between green bonds and conventional financial instruments is crucial to ensure a robust theoretical foundation encompassing various research areas. Sustainable finance, which is a critical domain within environmental finance, delves into concepts such as risk management, valuation models, and portfolio theory.
Green bonds are financial instruments that have been designed to fund environmentally friendly projects that align with broader sustainability goals. The study of green bonds within the context of European banks is grounded in the premise that financial institutions play a crucial role in fostering environmental responsibility through their investment choices. The nexus also taps into theories of financial market integration. Understanding their relationship with European banks becomes imperative as green bonds become increasingly integrated into financial markets.
This framework, as articulated by Clark et al. (2015), provides insights into the analysis of financial aspects related to bonds and sustainable activities [10]. Additionally, this examination has notable implications for corporate social responsibility. Margolis et al. (2003) highlight that investments in environmentally friendly projects can strategically benefit firms in this context [11]. Taking an alternate viewpoint, institutional isomorphism, a subset within institutional theory, posits that organizations adopt structures and practices to align with institutional norms and expectations [12]. In examining the connections with banks, we explore how these financial institutions adhere to or deviate from the institutional pressures associated with sustainable finance.
Herein, we align our approach with market-based theories, positing that information regarding green bonds, encompassing insights into sustainable practices, is mirrored in the stock prices of the banking industry. This perspective sheds light on market perceptions and reactions, as elucidated by the efficient market hypothesis articulated by Fama (1970) [13].
Moreover, many recent papers have conducted empirical analyses to tackle these issues, with numerous findings corroborating our methodology. To establish the links between the green bond market and other financial markets (the possibility that a particular market may influence or be influenced by other markets; in other words, receiving or transmitting risks), Pham’s study is employed as a reference [14]. A multivariate GARCH model is applied to the S&P green bond index between April 2010 and April 2015, and the results show that a shock generated by the conventional bond market spread through the green bond market, but the effect was variable over time.
Based on dynamic correlations for the different time intervals and the error decomposition of a multivariate vector autoregressive vector model for the European and American markets, it has been shown that state and corporate bonds influence green bonds. Still, the effects that are transmitted from green bonds are negligible [15]. At the same time, the link between high-yield corporate bonds, stocks, and energy assets is weak.
The asymmetric connection between green bonds and other assets was analyzed [16] using the methodological approach developed by Diebold and Yilmaz [17] and Baruník and Křehlík [18]. The results showed that of the commodities, only gold and silver tend to have a strong link with green bonds. In the short term, connections with oil-based assets are weak, but, in the long run, they become strong. Conversely, the transmission of the effects of positive returns is stronger in the short term, while the adverse effects are maintained over both periods but are more pronounced in the short term.
Quantile analysis has become increasingly popular as a way to establish the possible links between various assets. Data from the US green bond index for December 2013 to January 2019 showed a significant two-way causality between the oil price and the green bond index and a one-way relationship from geopolitical risk to the green bonds index for low quantiles [19]. For a similar period (April 2013–December 2019), another author provides evidence on the two-way causality between green bonds and other assets (Bitcoins, S&P 500, the clean energy index, the GSCI commodity index, and the CBOE volatility index) in different quantities [20]. Similarly, quantile regression was employed to establish a connection between the environmental attention index of cryptocurrencies and green bonds, among other asset classes. The authors concluded that a positive influence exists, although it is not statistically significant, between the S&P green bond index and the attention index [21].
The onset of the COVID-19 outbreak substantially impacted the financial markets, particularly in the initial months. Nevertheless, green investments exhibited a degree of resilience during this period [22], and green portfolios had better hedging [23]. Specifically addressing green bonds, other analyses revealed that they function within the system as net transmitters of spillovers in the short term and as net receivers in the long term [24]. When compared to other European conventional bonds, green bonds exhibit elevated risk exposure and diminished resilience to financial distress; they demonstrate a heightened sensitivity to favorable market conditions, garnering more substantial gains [25]. In the broader landscape of economic uncertainty in the US and EU, green bonds demonstrate heightened resilience to specific shocks compared to their conventional counterparts. At the same time, recent findings highlight the enhanced resilience of green bonds amid the particular challenges posed by the COVID-19 crisis [26]. The results suggest an increase in green bond uptake after the COVID-19 pandemic, and their significant evolvement compared to pre-COVID-19 figures proves the green bonds’ popularity.
According to our information, no studies have been carried out involving the European banking sector, which is an essential segment for lending in this region of operation. In addition to specialized indices on American and European companies, we also focus this study on the links of the green bond index with a representative index for companies in emerging countries. Moreover, our analysis also considers those sectors of activity not explored from this point of view: real estate, IT, and utilities. At the same time, given the consistent evolution of the green bond sector, we complement the findings of existing studies with an analysis based on recent information. The research methodology is based on a new approach developed by Han et al. [9] that was further implemented in other papers [27,28,29,30].

3. Research Methodology

Considering the nonlinear dynamics governing international financial markets, our study employs a methodology that has been carefully chosen to unveil the intricate relationship between green bonds and the banking system, particularly exploring connections that may emerge during challenging market conditions. Our analyses take two forms: first, examining the link between a green bond index and the banking industry, and second, investigating relationships between the green bond index and various other financial market indices. Another objective involves comparing the former with the latter to illustrate the extent of differentiation within the banking industry.
The data used in this analysis were obtained from Bloomberg and have a daily frequency, covering July 2014 to January 2021, i.e., 1705 observations.
Green bonds were analyzed using the S&P green bond index, representing the global green bond market. It is the first index of this kind, launched on 31 July 2014, and the values corresponding to the previous period do not represent the actual performance. However, they are estimated based on the methodology for calculating the index in force at launch. Only those bonds for which the income is used to finance green projects are included in the index and it represents a fundamental benchmark in academic research. Although this is a global index, according to the European Court of Auditors (2021), 48% of the green bonds issued in 2020 were European [31].
To compare the green bond market’s relationship with traditional financial markets, we considered various categories of stock market indices. Specifically, we included a bank index in our analysis to examine its correlation with the European lending sector.
Representative of the US stock market, the S&P 500 Composite Index contains listed companies with a strong focus on stock market capitalization and measures overall market performance. It was included in the current analysis to look at the relationship between green bonds and the general US stock market. Similarly, the STOXX Europe 600 index, which contains 600 European companies, was chosen for European companies. In addition, the STOXX Europe 600 Technology Index, one of the 20 major sectors included in the STOXX 600, was introduced in the analysis. STOXX Ltd. is the name of the company that designs these two indices and belongs to Qontigo, part of Deutsche Börse Group with headquarters in Eschborn.
We also included a group of indices developed by MSCI (Morgan Stanley Capital International): MSCI World (covering 1558 companies from 23 developed countries) and MSCI Europe (covering 430 companies and 15 developed European countries) which are representative indices for world and European companies, respectively; MSCI EM 50 comprises 50 companies from emerging countries, while MSCI World Real Estate comprises 95 companies from 23 developed real estate countries; MSCI World IT comprises 188 IT companies from 23 developed countries; MSCI World IT Services, MSCI World H/C Tech, and MSCI World Utilities comprises 84 companies from 23 developed countries. This orientation to the IT domain is partly motivated by increased research interest concerning green computing [32].
Our research hypothesis is that there is a connection between the European banking system and the green bond index. Testing this hypothesis is complemented by identifying the various facets of this connection and other types of dependence on the relevant stock market indices. To determine the extent to which the green bond index is connected to other stock indices when the market is experiencing extreme positive or negative movements, we will use a cross-quantilogram methodology developed by Han et al. [9]. According to the authors, this methodology allows for the measurement of quantile dependence between two series. The method has been applied in several areas but has also been accepted and included among the possibilities for analyzing green funding. Thus, we mention some recent examples of its use: to analyze the correlation and dependence between stocks, based on the energy produced from renewable resources and other asset classes [33]; to compare hedging opportunities with green bonds before and after COVID-19 in the US and China [34]; to analyze the dependence between green bonds and green stocks [35]; to highlight the asymmetric dependence between green bonds and other asset classes or between green bonds and commodities [36].
In practice, this methodology makes it possible to quantify the power and duration of the transmission of effects between markets in several categories of profitability distributions (such as extremely negative, central, or extremely positive observations) and for many lags.
To assess the appropriateness of this method, we employed the Lumsdaine–Papell (1997) test [37]. Under the null hypothesis, this test assumes the absence of a structural break in the time series. The results of this test across all the time series indicated that there was no evidence supporting the presence of structural breaks.
We will consider the stationary time series yit, where i represents the yields of the asset or index and t represents the time and i = 1, 2, t = 1, ..., T. The distribution and density functions of the yit series are Fi(·), respectively fi(·). The range of quantiles we want to analyze is α, αi ∈ (0, 1), and the function corresponding to the quantiles is:
q i t α i = i n f ϑ : F i ϑ α i .
For two events y 1 t q 1 t α τ 1 and y 2 t k q 2 t k α 2 , where k represents the length of the lag and k = ±1, ±2, ...., the cross-quantilogram will be:
ρ τ k = E   φ α 1 y 1   t q 1   t α 1 φ α 2 y 2   t k q 2   t k α 2 E φ α 1 2 y 1   t q 1   t α 1   E   φ α 2 2 y 2   t k q 2   t k α 2
where φ a u   1 u < 0 a represents the quantile-hit process. The cross-quantilogram captures the dependence between the two series for different quantiles and remains unchanged in the case of the strictly monotone transformations applied to both series, such as, for example, the logarithmic transformation. In the case of these two events, if ρ τ k = 0 , there is no cross-dependence.
Our estimation procedure will rely on the estimation of the quantiles q i * ( α i ) , and, henceforth, the computation of the sample cross-quantilogram, as formulated by Han et al., such that:
ρ α * k = t = k + 1 T φ α 1 y 1   t q 1   t * α 1 φ α 2 y 2   t k q 2   t k * α 2   t = k + 1 T φ α 1 2 y 1   t q 1   t * α 1   k = t + 1 T φ α 2 2 y 2   t k q 2   t k * α 2   .
The value of ρ α * k quantifies the dependence in quantiles of the two time series and is bounded to the interval [ 1 ,   1 ] .
The statistical significance of ρ τ k was verified using the Ljung–Box test developed by Han et al. (2016) [9] in the calculation:
Q α * p = T ( T + 1 ) k = 1 p ρ α * 2 ( k ) T k
while to approximate the test distribution according to the null hypothesis, a stationary bootstrap known as a block bootstrap, as defined by Politis and Romano (1994) [38], is used.
By analyzing how ρτ(k) varies according to the lag k, we can identify how the cross-dependence, depending on the cross-quantilogram between two markets, varies for different time horizons, thus highlighting the extent and duration of the dependence.

4. Results

An analysis of the importance of green bonds was carried out from two perspectives: on the one hand, we considered the study of the link between the green bond index and the European banking system, and on the other hand, our investigation aimed at estimating the links between this index and other stock indices that have the potential to be influenced or to have an impact on the financing of green projects.
The initial analysis stage, namely, the link between the European banking system and the green bond index, involved conducting investigations from two perspectives.
The first perspective aimed to create an index of the European banking market, representing our analysis. Companies classified as banking institutions and that were included in the STOXX 600 index were identified according to the GICS (Global Industry Classification Standard—MSCI) classification. The banking institutions thus recognized and used in the analysis are presented in Table 1.
Based on the logarithmic returns calculated for each targeted company, an index was constructed to reflect the dynamics of the entire group by achieving a simple arithmetic mean of these returns on each trading day. The index thus obtained, from now on referred to as the European banking index, was used to analyze the link between the European banking system and the evolution of the green bond index.
Two directions of dependence were investigated, from the bank index to the green bond index and vice versa. Both investigations are based on the cross-quantilogram methodology presented above. The quantile levels of 0.1, 0.5, and 0.9 were chosen, for which the cross-quantilograms on each pair were calculated to ascertain three types of dependencies: the dependencies according to the pair of quantiles of 0.1 (for the index of green bonds) and 0.1 (for the lags related to the bank index), followed by those for the pairs 0.5 and 0.5m and ultimately those concerning the pair for 0.9 by 0.9.
Figure 1 shows, at the top, the values of the quantilograms of the European banking index, calculated for the three levels for up to 20 lags. We can observe that these values exceed the levels drawn by the two dotted lines, which represent the 95% confidence intervals calculated for each lag via the stationary bootstrap methodology, especially in the case of the first lags for the pair 0.5 by 0.5, which means that the banking system transmits to the green bond index.
At the bottom, the graph shows the Ljung–Box test (portmanteau test) values and their critical values for each lag. We note that we have no statistical significance for these levels of quantiles and these analyzed pairs.
Another dimension of the analysis captures the quantified dependencies using quantilograms for several possible pairs. Figure 2 shows on the left the set of all dependencies between the green bond index and the first lag of the European banking index, for all possible combinations of quantiles corresponding to levels 0.1, 0.2, …, 0.9. In this case, we notice that we registered statistical significance for the pair of 0.4 by 0.4. The right side of the graph shows the dependency situation for the lower quantile values of 0.008, 0.009, 0.01, and 0.02, respectively.
We observe many significant combinations, particularly in the very high quantiles, in contrast to other scenarios. This outcome indicates that the connections are crucial during challenging market conditions when the banking industry and green bonds experience adverse shocks.
We can conclude that the European banking index generally transmitted contagion to the green bond index for the sample covering the whole period analyzed, due to the statistical significance (indicated by the presence of the sign *) for the combinations of pairs 0.03, 0.04, and 0.05, in the case of bank index lags, and for values 0.01 and 0.02 for the green bond index.
A final step in our analysis shows the dynamics of cross-quantilogram values for the dependence of the green bond index on the European banking index (at a lag). The calculation indicates that the calculated values are mainly in the evolution within the 95% confidence interval (calculated with the stationary bootstrap method), except for 2021, for all three pairs of quantiles, and 2020, for the pair 0.5 by 0.5 (Figure 3). We note the impact of the COVID-19 pandemic on the identified dependencies.
The reverse of the link between the European banking index and the green bond index is then investigated further. Thus, Figure 4 shows the dynamics of quantilograms up to 20 lags of the green bond index. We can conclude that there is no statistical significance for the quantilogram values represented in these graphs.
In the same register, in the case of an analysis on several pairs of quantile levels, Figure 5 shows that the green bond index has no statistically significant quantilogram values, neither for the large pairs (the left side of the chart) nor for the small ones.
In dynamics, when presenting the values of the quantilograms for each year, we notice that the values leave the range of 95% for all pairs in 2021 and in 2020 for the pair denoted by 0.5 by 0.5 (Figure 6).
The second perspective of the current analysis of the connection between the financial system and the green bond index consisted of performing calculations aimed at deriving the set of all pairs of quantiles that can be formed with the set containing the values from 0.1, 0.2, … 0. 9. Figure A1 and Figure A2 show the values of these quantilograms for the situations related to the first and second lag of the green bond index, respectively.
Following this analysis, we can report that only 26 situations were detected in which the values obtained were significantly different from zero; out of a total of 2268 combinations that were analyzed, 1.2% of the cases were different (in comparison with the 95% confidence intervals obtained by using the stationary bootstrap methodology), all appearing only in the case of the single lag analysis.
We also performed an analysis of a scenario with quantiles smaller than 0.1 for all the combinations that can be formed with the set {0.008, 0.009, 0.01, 0.02}, both in the case of a lag and in a situation with two lags. No statistical significance was identified for any of these cases.
In terms of the connection from each bank to the green bond index, we obtained 72 scenarios in which the cross-quantilograms were statistically significantly different from zero, comprising approximately 3.2% of the total of the 2268 scenarios analyzed. We found that there was more than double the number compared to the previous scenario. For the second lag, seven cases with statistically significant values were found. A graphic representation is presented in Figure A3 and Figure A4.
The second stage of the analysis aimed to identify the links between other stock indices and the green bond index. The grouping of these indices is presented in the first part of Section 3. Following the analysis of this direction, it was found that the results obtained so far show that the direction of transmission of effects is from financial institutions to the green bond index, rather than vice versa. We observed many scenarios (with combinations of levels corresponding to the calculated quantiles) for which the crossed quantilograms produced statistically significant values according to the stationary bootstrap criterion (Figure A5). Conversely, the significant results were much less marked from the green bond index to the other stock indices. The obtained results are graphically represented in Figure A6.

5. Discussion

The cross-quantilogram approach has gained considerable traction in the research discourse about green bonds. As outlined in the literature, this methodology is adept at discerning the directional nature of relationships, particularly in the extreme tails. However, it is essential to acknowledge a prominent critique of the approach, which centers on its limitations in effectively addressing volatility clustering [34]. On the other hand, GARCH-type models, which also represent a preferred method in green bond research, have other limitations, including linear provision for assets and Gaussian supposition for distribution [39]. Another employed copula-based framework method involves the variance-covariance matrix adjective [34].
The results obtained (although they consider other types of financial assets) align with those results obtained previously and reported in the literature. Thus, we found that other assets influence green bonds, but the effects transmitted from green bonds are negligible. An earlier study based on the GARCH approach evidenced that shocks are transmitted from the conventional bond market to the green one (Pham, 2016) [14]; our results support this observation and show the continuity of the feature a few years later. Our findings are similar to those of Reboredo, Ugolini, and Aiube [15], in that the effects transmitted from green bonds are negligible.
The significance of the COVID-19 period in elucidating the interconnections between the bank index and green bonds was established through a meticulous analysis of the sampled data beginning from March 2020 until January 2021, i.e., 226 observations. We conducted a cross-quantilogram analysis for both lag 1 and lag 2, in which we investigated both the direction of influence from the lags of the green bond index toward the bank index and vice versa (Figure A7).
Our results show that during the pandemic period, for lag 1, some connections were present from the green bond index toward the bank index but only showed in the large values of the quantiles, while we acknowledge that there was very low dependence for both the low quantiles and lag 2. Interestingly, the other direction (from the bank index to the green bond index) is more significant at lag 1, especially for the lower quantiles, meaning there is a strong relation in the tails of the distributions of these two indices. We can consider this as substantial evidence that extreme values of the green bond index tend to be influenced by the extremes of the bank index. Previous investigations have revealed similar findings [36]. These findings hold significance for diversification strategies and risk management, given the novelty of green bonds as instruments in financial markets [40]. Furthermore, these results bear significance for facilitating a seamless green transition as the success of such a move relies on the stability and resilience of the financial market [41].

6. Conclusions

This study’s main objective was to evaluate the links that green bonds develop with the financial sector using a modern approach based on cross-quantilograms. The first stage of the analysis, namely, the relationship between the European banking system and the green bond index, was analyzed from two perspectives: from the bank index to the green bond index and vice versa. We can conclude that the European banking index generally transmitted contagion to the green bond index for the sample covering the analyzed period at low quantile values. We noted the impact of the COVID-19 pandemic on the dependencies identified in the dynamic cross-correlation chart. For the inverse of the link between the European banking index and the green bond index, we found that the green bond index did not show statistically significant quantilogram values for either large or small pairs. In terms of dynamics, regarding this sense of connection, we noticed that the values left the range of 95% for all pairs in 2021 and 2020 for the pair 0.5 by 0.5.
In the subsequent phase of our comprehensive analysis, our focal point shifted toward elucidating the intricate connections between the green bond index and a spectrum of other prominent stock indices. Through a meticulous examination of these interrelationships, our study has yielded compelling insights that delineate a discernible trend in the effects of transmission. Notably, the results thus far underscore a consistent and noteworthy observation: the prevailing direction of effect transmission manifests from the various stock market indices to the green bond index, rather than exhibiting a reciprocal influence. This asymmetry in the transmission dynamics suggests a nuanced and pivotal relationship, wherein the broader movements and fluctuations within conventional stock markets exert a discernible impact on the behavior and performance of the green bond index. This finding further explores the underlying mechanisms governing this unidirectional influence. Potential factors contributing to this observed dynamic warrant careful consideration, including market sentiment, risk perceptions, and broader macroeconomic conditions. Understanding these intricate linkages is imperative for stakeholders such as investors, policymakers, and financial analysts, as it sheds light on the nuanced interplay between traditional financial markets and the burgeoning domain of environmentally conscious investments.
Our findings suggest that during the pandemic period, for lag 1, there are some connections from the green bond index toward the bank index for the large values of the quantiles, while we noticed very low dependence for both the low quantiles and lag 2. From the bank index to the green bond index, we found a strong relationship in the tails of the distributions of these two indices. Our work substantiates the finding that extreme values of the green bond index tend to be influenced by the extremes of the bank index.
One of the limitations of this empirical study is that the analysis is confined to the listed companies, but this represents one of the standard approaches in the literature. Another limitation is the scant availability of information on green bonds; available data are scarce, and when they are found, they often cannot be compared due to a lack of globally accepted standards. Over time, transparency in reporting and the better collection of background information will allow for the development of research on the sub-categories of green bonds. Although the green bond market is relatively new, some special features are documented in the literature that can foster the emergence of some further research directions. Following the results of Fatica, Panzica, and Rancan, we can detail our analysis regarding different types of emitters, as supranational institutions and corporations display a premium compared with financial institutions [42]. Future research directions can be oriented toward behavioral economics to understand the reasons, the frequency and amplitude of behavioral changes, and the degree of acceptance of novelties or changes in perceptions.
The development of the green finance field brings changes in regulations, individual attitudes, and society as a whole. Moreover, the introduction of specific indicators that incorporate environmental criteria could help in measuring the results of economic activity. However, “green” does not mean risk-free instruments and knowledge of the links and influences received and transmitted by such instruments is necessary for economic policymakers, institutions that issue such securities, and, last but not least, investors to make well-informed decisions.
The concept of green financing is an increasing part of current discourse; the principles of green finance and best practice are assimilated on a larger scale, thus demonstrating the field’s growing importance and recognizing the need for such an approach. This study comprehensively explains the complex interplay between green bonds and European banks, contributing to sustainable finance and financial economics.

Author Contributions

Conceptualization, I.L. and R.L.; methodology, I.L.; software, R.L.; validation, A.C.; formal analysis, I.L.; investigation, A.C.; resources, R.L.; data curation, I.L.; writing—original draft preparation, I.L. and R.L.; writing—review and editing, I.L. and A.C.; visualization, R.L.; supervision, I.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data was obtained from Datastream and Bloomberg, available through the Bucharest University of Economic Studies.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Figure A1. Cross-quantilogram values for each bank with the first lag of the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Figure A1. Cross-quantilogram values for each bank with the first lag of the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a1
Figure A2. Cross-quantilogram values for each bank with the second lag of the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Figure A2. Cross-quantilogram values for each bank with the second lag of the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a2
Figure A3. Cross-quantilogram values for the first lag of each bank with the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Figure A3. Cross-quantilogram values for the first lag of each bank with the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a3
Figure A4. Cross-quantilogram values for the second lag of each bank with the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Figure A4. Cross-quantilogram values for the second lag of each bank with the green bond index. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a4
Figure A5. Cross-quantilogram values from the first lag of a group of stock indices to the green bond index (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Figure A5. Cross-quantilogram values from the first lag of a group of stock indices to the green bond index (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a5aEnergies 16 07974 g0a5b
Figure A6. Cross-quantilogram values from the first lag of the green bond index to each stock index in the chosen group (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Figure A6. Cross-quantilogram values from the first lag of the green bond index to each stock index in the chosen group (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a6aEnergies 16 07974 g0a6b
Figure A7. Cross-quantilogram values during the pandemic (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Figure A7. Cross-quantilogram values during the pandemic (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g0a7aEnergies 16 07974 g0a7b

References

  1. Caldecott, B.; McDaniels, J. Financial Dynamics of the Environment: Risks, Impacts, and Barriers to Resilience. 2014. Available online: https://wedocs.unep.org/handle/20.500.11822/9697 (accessed on 17 November 2023).
  2. Lupu, R.; Călin, A.C.; Zeldea, C.G.; Lupu, I. A Bayesian Entropy Approach to Sectoral Systemic Risk Modeling. Entropy 2020, 22, 1371. [Google Scholar] [CrossRef]
  3. Lupu, I. European Stock Markets Correlations in a Markov Switching Framework. Rom. J. Econ. Forecast. 2015, 18, 103–119. [Google Scholar]
  4. Sternberg, T. Chinese Drought, Bread, and the Arab Spring. Appl. Geogr. 2012, 34, 519–524. [Google Scholar] [CrossRef]
  5. Lupu, R.; Călin, A.C.; Zeldea, C.G.; Lupu, I. Systemic Risk Spillovers in the European Energy Sector. Energies 2021, 14, 6410. [Google Scholar] [CrossRef]
  6. Deschryver, P.; de Mariz, F. What Future for the Green Bond Market? How Can Policymakers, Companies, and Investors Unlock the Potential of the Green Bond Market? J. Risk Financ. Manag. 2020, 13, 61. [Google Scholar] [CrossRef]
  7. Balcerzak, A.P. Europe 2020 Climate Change and Energy Objectives in EU-15. In Proceedings of the 11th International Days of Statistics and Economics, Prague, Czech Republic, 6–7 December 2018. [Google Scholar]
  8. Jones, R.; Baker, T.; Huet, K.; Murphy, L.; Lewis, N. Treating Ecological Deficit with Debt: The Practical and Political Concerns with Green Bonds. Geoforum 2020, 114, 49–58. [Google Scholar] [CrossRef]
  9. Han, H.; Linton, O.; Oka, T.; Whang, Y.-J. The Cross-Quantilogram: Measuring Quantile Dependence and Testing Directional Predictability between Time Series. J. Econom. 2016, 193, 251–270. [Google Scholar] [CrossRef]
  10. Clark, G.L.; Feiner, A.; Viehs, M. From the Stockholder to the Stakeholder: How Sustainability Can Drive Financial Outperformance. 2015. Available online: https://papers.ssrn.com/abstract=2508281 (accessed on 21 November 2023).
  11. Margolis, J.; Walsh, J. Misery Loves Companies: Rethinking Social Initiatives by Business. Adm. Sci. Q. Admin Sci. Quart. 2003, 48, 268–305. [Google Scholar] [CrossRef]
  12. DiMaggio, P.J.; Powell, W.W. The Iron Cage Revisited Institutional Isomorphism and Collective Rationality in Organizational Fields. In Economics Meets Sociology in Strategic Management; Baum, J.A.C., Dobbin, F., Eds.; Advances in Strategic Management; Emerald Group Publishing Limited: Bingley, UK, 2000; Volume 17, pp. 143–166. ISBN 978-1-84950-051-7. [Google Scholar]
  13. Fama, E.F. Efficient Capital Markets: A Review of Theory and Empirical Work. J. Financ. 1970, 25, 383–417. [Google Scholar] [CrossRef]
  14. Pham, L. Is It Risky to Go Green? A Volatility Analysis of the Green Bond Market. J. Sustain. Financ. Invest. 2016, 6, 263–291. [Google Scholar] [CrossRef]
  15. Reboredo, J.C.; Ugolini, A.; Aiube, F.A.L. Network Connectedness of Green Bonds and Asset Classes. Energy Econ. 2020, 86, 104629. [Google Scholar] [CrossRef]
  16. Naeem, M.A.; Nguyen, T.T.H.; Nepal, R.; Ngo, Q.-T.; Taghizadeh–Hesary, F. Asymmetric Relationship between Green Bonds and Commodities: Evidence from Extreme Quantile Approach. Financ. Res. Lett. 2021, 43, 101983. [Google Scholar] [CrossRef]
  17. Diebold, F.X.; Yılmaz, K. On the Network Topology of Variance Decompositions: Measuring the Connectedness of Financial Firms. J. Econom. 2014, 182, 119–134. [Google Scholar] [CrossRef]
  18. Baruník, J.; Křehlík, T. Measuring the Frequency Dynamics of Financial Connectedness and Systemic Risk. J. Financ. Econom. 2018, 16, 271–296. [Google Scholar] [CrossRef]
  19. Lee, C.-C.; Lee, C.-C.; Li, Y.-Y. Oil Price Shocks, Geopolitical Risks, and Green Bond Market Dynamics. N. Am. J. Econ. Financ. 2021, 55, 101309. [Google Scholar] [CrossRef]
  20. Thai Hung, N. Quantile Dependence between Green Bonds, Stocks, Bitcoin, Commodities and Clean Energy. Econ. Comput. Econ. Cybern. Stud. Res. 2021, 55, 71–86. [Google Scholar] [CrossRef]
  21. Hassan, M.K.; Hasan, M.B.; Halim, Z.A.; Maroney, N.; Rashid, M.M. Exploring the Dynamic Spillover of Cryptocurrency Environmental Attention across the Commodities, Green Bonds, and Environment-Related Stocks. N. Am. J. Econ. Financ. 2022, 61, 101700. [Google Scholar] [CrossRef]
  22. Liu, X.; Bouri, E.; Jalkh, N. Dynamics and Determinants of Market Integration of Green, Clean, Dirty Energy Investments and Conventional Stock Indices. Front. Environ. Sci. 2021, 9, 575. [Google Scholar] [CrossRef]
  23. Akhtaruzzaman, M.; Banerjee, A.K.; Ghardallou, W.; Umar, Z. Is Greenness an Optimal Hedge for Sectoral Stock Indices? Econ. Model. 2022, 117, 106030. [Google Scholar] [CrossRef]
  24. Mensi, W.; Vo, X.V.; Ko, H.-U.; Kang, S.H. Frequency Spillovers between Green Bonds, Global Factors and Stock Market before and during COVID-19 Crisis. Econ. Anal. Policy 2023, 77, 558–580. [Google Scholar] [CrossRef]
  25. Cicchiello, A.F.; Cotugno, M.; Monferrà, S.; Perdichizzi, S. Credit Spreads in the European Green Bond Market: A Daily Analysis of the COVID-19 Pandemic Impact. J. Int. Financ. Manag. Account. 2022, 33, 383–411. [Google Scholar] [CrossRef]
  26. Contractor, D.; Balli, F.; Hoxha, I. Market Reaction to Macroeconomic Anouncements: Green vs Conventional Bonds. Appl. Econ. 2023, 55, 1637–1662. [Google Scholar] [CrossRef]
  27. Baumohl, E.; Lyocsa, S. Directional Predictability from Stock Market Sector Indices to Gold: A Cross-Quantilogram Analysis. Financ. Res. Lett. 2017, 23, 152–164. [Google Scholar] [CrossRef]
  28. Jiang, H.; Su, J.-J.; Todorova, N.; Roca, E. Spillovers and Directional Predictability with a Cross-Quantilogram Analysis: The Case of U.S. and Chinese Agricultural Futures. J. Futures Mark. 2016, 36, 1231–1255. [Google Scholar] [CrossRef]
  29. Todorova, N. The Intraday Directional Predictability of Large Australian Stocks: A Cross-Quantilogram Analysis. Econ. Model. 2017, 64, 221–230. [Google Scholar] [CrossRef]
  30. Zhou, Z.; Jiang, Y.; Liu, Y.; Lin, L.; Liu, Q. Does International Oil Volatility Have Directional Predictability for Stock Returns? Evidence from BRICS Countries Based on Cross-Quantilogram Analysis. Econ. Model. 2019, 80, 352–382. [Google Scholar] [CrossRef]
  31. European Court of Auditors Special Report 22/2021: Sustainable Finance: More Consistent EU Action Needed to Redirect Finance towards Sustainable Investment. Available online: http://www.eca.europa.eu/en/Pages/Report.aspx?did=59378&TermStoreId=8935807f-8495-4a93-a302-f4b76776d8ea&TermSetId=49e662c4-f172-43ae-8a5e-7276133de92c&TermId=5f6589a2-5a2e-4ae8-8cc6-e05bf544b71f (accessed on 17 November 2023).
  32. Lis, A.; Sudolska, A.; Pietryka, I.; Kozakiewicz, A. Cloud Computing and Energy Efficiency: Mapping the Thematic Structure of Research. Energies 2020, 13, 4117. [Google Scholar] [CrossRef]
  33. Uddin, G.S.; Rahman, M.L.; Hedström, A.; Ahmed, A. Cross-Quantilogram-Based Correlation and Dependence between Renewable Energy Stock and Other Asset Classes. Energy Econ. 2019, 80, 743–759. [Google Scholar] [CrossRef]
  34. Guo, D.; Zhou, P. Green Bonds as Hedging Assets before and after COVID: A Comparative Study between the US and China. Energy Econ. 2021, 104, 105696. [Google Scholar] [CrossRef]
  35. Pham, L. Frequency Connectedness and Cross-Quantile Dependence between Green Bond and Green Equity Markets. Energy Econ. 2021, 98, 105257. [Google Scholar] [CrossRef]
  36. Pham, L.; Nguyen, C.P. Asymmetric Tail Dependence between Green Bonds and Other Asset Classes. Glob. Financ. J. 2021, 50, 100669. [Google Scholar] [CrossRef]
  37. Lumsdaine, R.L.; Papell, D. Multiple Trend Breaks And The Unit-Root Hypothesis. Rev. Econ. Stat. 1997, 79, 212–218. [Google Scholar] [CrossRef]
  38. Politis, D.N.; White, H. Automatic Block-Length Selection for the Dependent Bootstrap. Econom. Rev. 2004, 23, 53–70. [Google Scholar] [CrossRef]
  39. Wang, K.; Chen, Y.-H.; Huang, S.-W. The Dynamic Dependence between the Chinese Market and Other International Stock Markets: A Time-Varying Copula Approach. Int. Rev. Econ. Financ. 2011, 20, 654–664. [Google Scholar] [CrossRef]
  40. Huynh, T.L.D. When ‘Green’ Challenges ‘Prime’: Empirical Evidence from Government Bond Markets. J. Sustain. Financ. Invest. 2022, 12, 375–388. [Google Scholar] [CrossRef]
  41. Arif, M.; Hasan, M.; Alawi, S.M.; Naeem, M.A. COVID-19 and Time-Frequency Connectedness between Green and Conventional Financial Markets. Glob. Financ. J. 2021, 49, 100650. [Google Scholar] [CrossRef]
  42. Fatica, S.; Panzica, R.; Rancan, M. The Pricing of Green Bonds: Are Financial Institutions Special? J. Financ. Stab. 2021, 54, 100873. [Google Scholar] [CrossRef]
Figure 1. Cross-quantilogram analysis of the bank index lags and green bond index values. Source: the authors’ own calculations, based on Bloomberg data.
Figure 1. Cross-quantilogram analysis of the bank index lags and green bond index values. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g001
Figure 2. Cross-quantilogram analysis of the bank index lags and green bond index values for various quantile combinations (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Figure 2. Cross-quantilogram analysis of the bank index lags and green bond index values for various quantile combinations (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g002
Figure 3. Dependency dynamics for three pairs of quantiles between the green bond index and the first lag of the European banking index. Source: the authors’ own calculations, based on Bloomberg data.
Figure 3. Dependency dynamics for three pairs of quantiles between the green bond index and the first lag of the European banking index. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g003
Figure 4. Cross-quantilogram analysis between green bond index lags and European bank index values. Source: the authors’ own calculations, based on Bloomberg data.
Figure 4. Cross-quantilogram analysis between green bond index lags and European bank index values. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g004
Figure 5. Cross-quantilogram values for several pairs of levels and lag 1 of the European banking index for various quantile combinations (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Figure 5. Cross-quantilogram values for several pairs of levels and lag 1 of the European banking index for various quantile combinations (* prompts statistically significant). Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g005
Figure 6. Dependency dynamics for three pairs of quantiles between the green bond index and the first lag of the European banking index values for various quantile combinations. Source: the authors’ own calculations, based on Bloomberg data.
Figure 6. Dependency dynamics for three pairs of quantiles between the green bond index and the first lag of the European banking index values for various quantile combinations. Source: the authors’ own calculations, based on Bloomberg data.
Energies 16 07974 g006aEnergies 16 07974 g006b
Table 1. Banking institutions included in STOXX 600 and used in the analysis.
Table 1. Banking institutions included in STOXX 600 and used in the analysis.
Name of Banks
1AIB Group PLC
2Banco Bilbao Vizcaya Argentaria SA
3Banco BPM SpA
4Banco de Sabadell SA
5Banco Santander SA
6Bank of Ireland Group PLC
7Bank Polska Kasa Opieki SA
8Bankinter SA
9Barclays PLC
10BNP Paribas SA
11CaixaBank SA
12Commerzbank AG
13Credit Agricole SA
14Danske Bank A/S
15Erste Group Bank AG
16HSBC Holdings PLC
17ING Groep NV
18Intesa Sanpaolo SpA
19KBC Group NV
20Lloyds Banking Group PLC
21Nordea Bank Abp
22Powszechna Kasa Oszczednosci Bank Polski
23Raiffeisen Bank International AG
24Santander Bank Polska SA
25Skandinaviska Enskilda Banken AB
26Societe Generale SA
27Standard Chartered PLC
28Svenska Handelsbanken AB
29Swedbank AB
30UniCredit SpA
31Unione di Banche Italiane SpA
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lupu, I.; Lupu, R.; Criste, A. The Nexus between Green Bonds and European Banks: A Cross-Quantilogram Approach. Energies 2023, 16, 7974. https://doi.org/10.3390/en16247974

AMA Style

Lupu I, Lupu R, Criste A. The Nexus between Green Bonds and European Banks: A Cross-Quantilogram Approach. Energies. 2023; 16(24):7974. https://doi.org/10.3390/en16247974

Chicago/Turabian Style

Lupu, Iulia, Radu Lupu, and Adina Criste. 2023. "The Nexus between Green Bonds and European Banks: A Cross-Quantilogram Approach" Energies 16, no. 24: 7974. https://doi.org/10.3390/en16247974

APA Style

Lupu, I., Lupu, R., & Criste, A. (2023). The Nexus between Green Bonds and European Banks: A Cross-Quantilogram Approach. Energies, 16(24), 7974. https://doi.org/10.3390/en16247974

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop