ESG as risk factor

Dobrick, Juris; Klein, Christian; Zwergel, Bernhard Article — Published Version ESG as risk factor Journal of Asset Management Provided in Cooperation with: Springer Nature Suggested Citation: Dobrick, Juris; Klein, Christian; Zwergel, Bernhard (2025) : ESG as risk factor, Journal of Asset Management, ISSN 1479-179X, Palgrave Macmillan, London, Vol. 26, Iss. 1, pp. 44-70, https://doi.org/10.1057/s41260-024-00382-z This Version is available at: https://hdl.handle.net/10419/319247 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/ Vol:.(1234567890) Journal of Asset Management (2025) 26:44–70 https://doi.org/10.1057/s41260-024-00382-z S.I.: RISKS RELATED TOENVIRONMENTAL, SOCIAL ANDGOVERNMENTAL ISSUES (ESG) ESG asrisk factor JurisDobrick1 · ChristianKlein1· BernhardZwergel1 Revised: 14 May 2024 / Accepted: 30 October 2024 / Published online: 15 January 2025 © The Author(s) 2024 Abstract There are numerous risk factors in asset pricing models that have been identified over the years. In this paper, we address the question of whether factors constructed using ESG (Environmental, Social, Governance) scores could potentially meet the necessary requirements for risk factors in multifactor models. While numerous studies indicate that the ESG performance of firms could be financially material, the integration of ESG factors has so far not been fully evaluated. We pay particular attention to the problem of divergent scores across different rating providers and investigate whether the regression results of 4and 5-factor models converge. The evaluation is carried out with Fama–French and Carhart models, extended by an additional factor representing ESG, respectively. We find that there are ESG factors across all investigated rating providers that capture common-variation in stock returns over time, indicating that ESG should be considered in common asset pricing models. Keywords ESG· Portfolio management· Risk factors· ESG integration Introduction The financial performance of investments meeting certain ESG (Environmental, Social, Governance) criteria in comparison with such that do not has been investigated in multiple contributions (e.g., Derwall etal. 2005; Humphrey etal. 2012). Although the findings seem to differ, Friede etal. (2015), Wallis and Klein (2015) as well as more recently, Atz etal. (2020) demonstrate that the bulk of research in that field indicates an at least nonnegative relation between the consideration of ESG criteria and financial performance. Deviations from this observation are mainly explained by three different approaches: the error-in-expectations hypothesis, the shunned-stock hypothesis and the environmentalrisk hypothesis. The error-in-expectations hypothesis is used to explain a possible outperformance of ESG investments compared to conventional investments. It suggests that ESG investments can deliver abnormal returns, because positive ESG information, in contrast to negative ESG information, is not fully or ambiguously priced by investors when it becomes public (e.g., Derwall etal. 2011; Capelle-Blancard & Petit 2019; Krüger 2015). The shunned-stock hypothesis states that non-financial investor preferences lead firms with weaker ESG performances to earn abnormal returns (Hong & Kacperczyk 2009). Based on the work of Merton (1987) on segmented markets, it argues that the shunning of stocks/firms with weak ESG performances by norm-oriented investors leads to a systematic undervaluation and thus to higher risk-adjusted returns. Additionally, limited risk sharing among non-normoriented investors causes idiosyncratic risk and not just beta pricing to matter (Hong & Kacperczyk 2009). Finally, the environmental-risk hypothesis regards firms’ ESG performance as systematic risk for which investors demand compensation. The lower the ESG performance the higher the ESG risk of a firm (i.e., reputational or regulatory risk) and the higher the (expected) return. The non-consideration of ESG as systematic risk in standard asset-pricing models could therefore falsely lead to the indication ofan * Juris Dobrick [email protected] Christian Klein [email protected] Bernhard Zwergel b.zwer[email protected] 1 University ofKassel, Kassel, Germany 45ESG asrisk factor outperformance of low-ESG stocks compared with high ESG-stocks. While some studies provide initial evidence that an ESG risk factor is able to explain returns on ESG investments, there has not yet been a study that systematically looks at the explanatory power of ESG risk factors. This study addresses the latter of the underlying theories and asks whether there are risk factors related to ESG that should be incorporated into common asset-pricing models. Initially, we investigate this for the global market. In a robustness section, we check for such factors in other markets (i.e., North America and Europe). ESG-related risk factors are represented by classic zero-investment portfolios which might explain variation over time in stocks that are considered unsustainable (i.e., they have low ESG ratings) as well as such with high ESG ratings, which we refer to as sustainable here. By implementing the above, we add to the numerous risk factors that have been identified over the past decades which seem to help explain the cross section of expected returns (e.g., Harvey etal. 2016; Cochrane 2011), although the way they are linked to economic fundamentals remains for the most part unclear. Against this background we want to take up the objections made in connection with the “zoo of factors” (Cochrane 2011) with particular focus on too low significance levels leading to false positives (Harvey etal. 2016). Furthermore, we consider the requirements for robust factors by Beck etal. (2016), namely the grounding in a long and deep academic literature and the robustness across definitions. We address the robustness across definitions criteria by using ESG ratings of three different providers. We do so due to the findings of, e.g., Dorfleitner etal. (2015), Chatterji etal. (2016) and Berg etal. (2022) that ESG ratings differ. By constructing factors built upon three different raters, we hope to show that despite of the divergence between ESG raters, it is possible to construct risk factors on the basis of each of them, which explain common variation in returns of unsustainable and sustainable assets. If so, such factors could be used, e.g., in portfolio management to steerthe risk with respect to ESG. It is worth noting that the first requirement to risk factors made by Beck etal. (2016)—the grounding in a long and deep academic literature—is quite hard to account for in the context of ESG, since ESG considerations and the dissemination of rating data have unlike economic fundamental data been initiated as late as roughly two decades ago. From a theoretical standpoint, we argue that there is a sufficient amount of theory that suggests the existence of risk factors related to ESG. In fact, quite recently, Pástor etal. (2021) point out in an equilibrium model that from a perspective of risk, ESG stocks should yield lower returns in equilibrium due to their resilience to ESG-related shocks such as unexpected changes to climate change related regulation. This view is supported by Cornell (2020). Interestingly, it is argued that during transition periods when investor preferences are changing or when climate-related concerns increase, higher returns can be expected for sustainable assets since prices are driven upward (Pástor etal. 2021). Given these arguments, factor premiums related to ESG could tend in either direction depending on how overall demand is shifting in favor of sustainable assets during a given timespan, therefore making it hard to derive a universally valid conclusion related to the direction of such a factor premium under the assumption of rational pricing. Pástor etal. (2021) recognize this fact and stress the difficulty of disentangling positive alphas owed to shifts in investors’ tastes from alphas that are in fact due to some underlying risk component. Consequently, any findings in this paper are more of a snapshot of how markets have valued ESG during a given timespan than they are predictive of the direction future valuations will take. To this end, one might raise the question as to whether a factor which may not be priced is useful at all. Pástor etal. (2022) show that changing concerns related to climate change explain the bad performance of value stocks in the post-2010s, leading to a negative premium of the HML factor of Fama and French (1993). As this factor was originally developed in a way such that its premium is positive, i.e., value stocks are expected to outperform growth stocks, it is possible that this premium reverses. Hence, we argue that for a given timespan, which is relatively short, a factor premium can average out to zero or can shift its direction, which is why we see the existence of a factor premium in the context of ESG as secondary. To sum up, this paper joins the ranks of research related to the question of whether there are risk factors related to ESG that capture common variation in stock returns over time and which are consistent across definitions (i.e., across different rating providers). Initially, this is done for a global portfolio. In a robustness section, we extend the analysis to further markets to demonstrate factor robustness across different markets. Similar research in that field was conducted by Lioui (2018) who finds thatthe average return of a factor which is long in stocks with low ESG strengths and short in such with high ESG concerns is significantly different from zero. Pollard and Sherwood (2018) show that ESG premia provide excess returns, Oestreich and Tsiakas (2015) along with Görgen etal. (2020) show that factors related to carbon risk explain common variation in stock returns over time. In a similar vein, Hsu etal. (2023) propose a risk factor that controls for environmental political uncertainty. Ciciretti etal. (2017) find further evidence for the existence of an ESG premium, Jin (2018) underscores these findings for the US market and Hübel and Scholz (2018) find increased 46 J.Dobrick et al. explanatory power for models extended by an additional factor proxy for ESG. With our study, we contribute further important insights with respect to the relevance of such factors for common asset-pricing models. If there are systematic risk components related to ESG, we expect the produced loadings of ten decile portfolios built upon ESG data with ascending scores across deciles on our UMS (unsustainable minus sustainable) risk factor in time-series regressions to be positive (negative) for the lowest (highest) scores. For deciles that are somewhere in the middle between the lowest and highest deciles, we expect factor loadings on UMS that are rather insignificant since those deciles should be neither strongly influenced by negative shocks related to ESG risks nor should they be entirely unaffected. According to our knowledge, we are the first to consider ESG data by three different rating providers, different markets as well as several dimensions of ESG and therefore close an important research gap related to ESG and factor robustness in the sense of Beck etal. (2016). We find that our factors capture common-variation in stock returns over time and enhance the explanatory power of common asset pricing models. This applies to factors built with rating data of three different providers, across all geographies as well as across all ESG dimensions. Consequently, we argue that ESG-related factors should be incorporated into common asset pricing models to manage the risk related to ESG. The remainder of this paper proceeds as follows. Section "Data and methods" comprises the implementations regarding data and methods. Section "Results" provides descriptive statistics and the results of our regressionanalyses. Section "Robustness checks" provides robustness regression results for additional markets. Section "Conclusion" concludes. Data andmethods Data collection andpreparation For our ESG risk factor in the time-series regression, we depend on firm-specific ESG ratings. We obtain ratings from ASSET4, LSE Refinitiv (henceforth Refinitiv) and Moody’s Vigeo Eiris (henceforth Vigeo Eiris). ASSET4 uses publicly available and traceable sources such as websites, SEC filings, sustainability reports, media sources, and NGO reports to derive more than 700 nonfinancial firm-level data points. Every data point is the firmspecific expression of a single ESG-related characteristic. These data points are aggregated in several stages into 18 categories that cover general ESG themes within the E, S, and G pillars. ASSET4 generates a rating for each category. A rating ranges from 0 to 100 points, where 100 indicates a very strong ESG performance relative to other firms in the firm universe. Moreover, to obtain a firm’s overall ESG performance, we calculate ratings as equally weighted aggregation of the E, S, and G pillars (henceforth EWR). Refinitiv provides data on companies’ sector specific ESG performance based on reported data. Altogether, ESG performance is measured in 10 different main themes pertaining to distinct pillars of ESG, where—just as for ASSET4— scores range from 0 to 100 points with increasing ESG performance. The first ESG ratings by Refinitiv and ASSET4 can be obtained from 2002. Vigeo Eiris ESG ratings measure the degree to which firms take into account and manage material ESG factors. Firms with higher ESG ratings are better at managing relationships with their stakeholders on a scale from 0 to 100. To generate ratings, Vigeo Eiris analyzes and scores up to 38 distinct ESG criteria that are framed within 40 industry specific models. In each industry framework, the 38 generic ESG criteria are assigned a weight that reflects the sector specific materiality of the analyzed criterion. Each criterion has a defined set of so-called Principles of Action. These determine the active content of the analysis and articulate the actions that Vigeo Eiris would expect a high-performing firm to undertake in this dimension. These principles are derived from universally recognized norms and standards emanating from organizations such as the United Nations, the International Labour Organization, and the Organisation for Economic Cooperation and Development. Within the rating process, qualitative and quantitative data, management and performance data as well as self-reported and third-party data are used. Vigeo Eiris offers ESG ratings for the E, S and G pillar as well as a general ESG assessment for an international sample of firms for a time series starting in 2003. While these three rating providers calculate ESG ratings for the same ESG dimensions, i.e., the E, S, and G pillars as well as an aggregated ESG assessment, they differ in firm coverage. To ensure that potential deviations in factor loadings of the ESG risk factor for the respective rating provider do not arise due to different firm coverages but are attributable to the differences in rating methodologies, we run the regressions for each ESG rating provider based on the same set of firms. After constraining our firm sample to those firms that are contained in each initial set of the three agencies’ firm universes, we pair ESG data with financial data from Refinitiv such that we end up with a universe of roughly 4500 companies. Since Refinitiv ESG ratings measure a company’s ESG performance relative to its industry, we recalculate the scores following the methodology of Dyck etal. (2019) to finally obtain ratings data relative to our entire matched equity universe. Furthermore, to adjust for gaps in the time-series of Vigeo Eiris ESG ratings, we perpetuate each score to two years in the future if there are no scores 47ESG asrisk factor available. This adjustment and the way it is set out is chosen to prevent our sample from becoming too small and to thus avoid inconclusive results due to idiosyncratic characteristics in our portfolios. To underpin this approach, we assume that a company’s ESG score will not change too drastically from one year to another especially without the consideration of ESG controversies. With respect to ASSET4 ESG ratings, no further corrections are made. Thereafter, ES (Environmental and Social) as well as ESG (Environmental, Social, Governance) scores are calculated by taking the arithmetic mean of the respective pillars such that we finally obtain five ESG indicators (Environmental, Social, Governance, ES and ESG) to measure a company’s ESG performance. We obtain financial data from Refinitiv. To calculate returns, we take the total return prices for each company to account for course movements due to dividend payouts and stock splits. Following Ince and Porter (2006), further adjustments are made either to the returns themselves or to the underlying total return prices to ensure high similarity to CRSP (Center of Research in Security Prices) return data. Therefore, we exclude all penny stocks (stocks with a price equal to or below 1$) to preclude erroneous return data due to rounding issues. Furthermore, we correct for returns above 300% which reverse within one month and dismiss those returns which are equal to 0 for more than two months in a row. Finally, we remove those returns which are not available throughout every month starting in July of every year t to June of t+1 to prevent our portfolio returns from artificial volatility due to missing data. We drag market value data for every company contained in our matched universe at the end of June of every year t. This point in time is chosen in accordance with Fama and French’s (1993) methodology to account for the fact that the required accounting data to construct their value factor must be available at the time when the regressions are conducted. Since data on companies’ market value are available throughout the entire year, we obtain our respective data at the closest point in time with respect to our regressions. The SMB (small minus big), HML (High minus low) and WML (winners minus losers) factors for the Fama–French threeand the Carhart four-factor models are drawn from Kenneth French’s data library. We use the Fama–French developed factors for the regressions for our regressions with global portfolios. For the risk-free rates, we use one-month U.S. Treasury bills since this study is from the perspective of an U.S. investor. To ensure broad portfolio diversification for the dependent as well as the independent returns our timeseries for our regression for the global market starts in July 2007 and ends in June 2020. Accordingly, our time-series finally comprises 156 observations of monthly returns. Variable construction To construct our risk factor proxy related to ESG, we follow the methodologies of Fama and French (1993) and Görgen etal. (2020) and build value-weighted UMS (unsustainable minus sustainable) portfolios which are long in unsustainable and short in sustainable assets. We go long unsustainable stocks since those stocks’ expected returns should be larger than those of the sustainable part (see, e.g., Pástor etal. (2021)) and we want to choose a setting in which positive factor premiums can at least be expected. To adjust for the evidence that firms with high market capitalizations tend to have higher ESG ratings (see, for example, Drempetic etal. (2020), Humphrey etal. (2012) or Brammer etal. (2006)), we take the market value of every firm in June of every year t and split the data at its median into a small firm sample (S) and one big firm sample (B). We then separately sort the same stocks according to their particular ESG score of every year t-2 and construct one unsustainable (L) and one sustainable (H) subset of stocks for every year t by taking the 30th and 70th percentile as thresholds, respectively. Like in Fama and French (1993), these thresholds have no deeper economic intuition and are chosen arbitrarily. Thereafter, we match the resulting portfolios with those from the size splits. The lag in the ESG data of two years is used for the same reason as for which Fama and French (1993) lag their data with respect to their value factor, i.e., since rating providers need time to provide ESG ratings for a given year, a lag is introduced to ensure the availability of the ratings at the time of portfolio construction. Here, we take a conservative approach and assume that it takes more than six months until ESG ratings for a given year are made public. As robustness check, we also construct factors with ESG data of t-1. Finally, we build our UMS portfolio as the difference of the arithmetic means of S/L, B/L and S/H, B/H and calculate monthly returns for July of year t until June of t+1 after which the portfolios are reformed. This is done for every rating provider as well as for each ESG dimension leading to 5 x 3 different factor constructions. Consistent with Fama and French (1993), our proxies for the market portfolios are value-weighted portfolios which comprise all stocks that are contained in our matched universe with respect to the relevant market. Thereafter, we calculate monthly excess returns by subtracting the respective proxy for our risk-free rate from our calculated market return. Our dependent returns in the time-series regressions are excess returns of ten decile portfolios constructed with ascending ESG scores, whereby the first decile contains those stocks whose scores fall below the 10th percentile and so forth. The 48 J.Dobrick et al. decile portfolios are recalculated—just as it applies to the UMS portfolio allocations—by the end of June of every year t. Finally, we calculate monthly excess returns by subtracting our monthly data of our respective risk-free rate proxy from every monthly return observation of each decile portfolio. The next section initially presents our descriptive statistics related to our constructed variables, whereafter we present our results for the time-series regressions. Results Descriptive statistics Descriptive statistics and factor correlations are presented in Tables1 and 2 and present an overview of the variables used in our asset-pricing tests. The subsequent part of this section analyzes the magnitudes and significances of the variables for our investigated market. Our generated factor premiums do all not significantly differ from zero, except for the premium generated by the HML factor. All other independent variable factor premiums are insignificant. Interestingly, the factor premiums of the Governance factor premium are negative in absolute magnitude across all three ratings providers, whereby all other premiums are positive with a few exceptions. While the premiums generated with ASSET4 and Refinitiv ESG data are about the same magnitude of standardized units away from a distribution with a mean of zero, premiums generated with Vigeo Eiris rating data are smaller in magnitude adjusted for standard deviations. Although the obtained correlations depicted in Table2 seem high in absolute magnitude and are highly significant on some occasions, we infer this circumstance to be negligible for two reasons. First, the standard Fama–French or Carhart factors, respectively, are also highly correlated in terms of Pearson’s r, second, because the correlations are all below 0.7, the critical value when it comes to detecting multicollinearity in regression models. For our sample period we conclude that our generated factor premiums are overall rather Table 1 Summary statistics: world This table shows means, standard deviations and significance levels for the independent variables in the time-series regressions for the global market *, ** and *** denote significance levels for independent t-tests on the 10%, 5% and 1% levels, respectively Mean SD Market − 0.002 0.052 SMB − 0.001 0.015 HML − 0.004** 0.020 WML 0.003 0.037 ASSET4 UMSENV 0.002 0.013 UMSGOV − 0.003 0.024 UMSSOC 0.002 0.014 UMSES 0.002 0.014 UMSESG 0.001 0.017 Refinitiv UMSENV 0.001 0.014 UMSGOV − 0.002 0.021 UMSSOC 0.001 0.014 UMSES 0.002 0.014 UMSESG 0.001 0.015 Vigeo Eiris UMSENV 0.002 0.023 UMSGOV − 0.0005 0.025 UMSSOC 0.002 0.023 UMSES 0.001 0.022 UMSESG − 0.0003 0.021 Table 2 Correlations This table shows Pearson correlations of the independent variables in the time-series regressions for the global market *, ** and *** denote significance levels on the 10%, 5% and 1% levels, respectively Market SMB HML WML Market 1.000*** 0.191** 0.186** − 0.420*** SMB 0.191** 1.000*** − 0.032 − 0.059 HML 0.186** − 0.032 1.000*** − 0.436*** WML − 0.420*** − 0.059 − 0.436*** 1.000*** ASSET4 UMSENV − 0.068 0.180** − 0.459*** 0.177** UMSGOV − 0.532*** − 0.017 − 0.237*** 0.371*** UMSSOC − 0.164** 0.225*** − 0.590*** 0.373*** UMSES − 0.156* 0.195** − 0.548*** 0.329*** UMSESG − 0.464*** 0.088 − 0.558*** 0.477*** Refinitiv UMSENV − 0.020 0.226*** − 0.429*** 0.071 UMSGOV − 0.548*** − 0.040 − 0.276*** 0.393*** UMSSOC − 0.361*** 0.118 − 0.453*** 0.479*** UMSES − 0.168** 0.240*** − 0.499*** 0.200*** UMSESG − 0.387*** 0.142* − 0.541*** 0.467*** Vigeo Eiris UMSENV 0.261*** 0.266*** − 0.261*** − 0.095 UMSGOV − 0.302*** − 0.033 − 0.212*** 0.270*** UMSSOC 0.016 0.232*** − 0.405*** 0.045 UMSES 0.190** 0.270*** − 0.305*** − 0.072 UMSESG 0.030 0.215*** − 0.328*** − 0.045 49ESG asrisk factor Table 3 ASSET4 world: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 0.995*** 1.049*** 1.029*** 1.031*** 1.038*** 1.054*** 1.005*** 0.961*** 1.029*** 0.978*** (0.029) (0.013) (0.022) (0.021) (0.020) (0.021) (0.019) (0.031) (0.026) (0.023) SMB − 0.005 0.138*** 0.047 0.180*** 0.168*** 0.095 0.008 − 0.014 − 0.126*** − 0.251*** (0.062) (0.048) (0.072) (0.064) (0.062) (0.060) (0.051) (0.047) (0.042) (0.031) HML − 0.021 − 0.003 − 0.025 − 0.043 0.013 0.006 0.078* 0.184*** − 0.055 0.106** (0.055) (0.051) (0.043) (0.063) (0.041) (0.059) (0.045) (0.051) (0.038) (0.049) WML − 0.003 − 0.028 − 0.071 − 0.050 0.019 − 0.076*** − 0.037 0.089*** 0.005 0.042*** (0.033) (0.020) (0.046) (0.033) (0.022) (0.023) (0.023) (0.031) (0.019) (0.013) UMS 0.492*** 0.565*** 0.546*** 0.113 0.240*** 0.242*** 0.171*** − 0.113 − 0.258*** − 0.307*** (0.082) (0.081) (0.071) (0.074) (0.056) (0.085) (0.065) (0.073) (0.049) (0.048) Constant 0.001 0.0005 − 0.001 0.002* − 0.0002 − 0.001* − 0.001 − 0.001 − 0.001 0.0002 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 292 286 289 288 289 289 288 289 289 288 Adjusted R20.958*** 0.966*** 0.954*** 0.964 0.972*** 0.968*** 0.977*** 0.972** 0.977*** 0.979*** Governance Market 0.938*** 0.970*** 1.012*** 1.106*** 1.156*** 1.100*** 1.045*** 0.977*** 0.977*** 0.900*** (0.027) (0.023) (0.032) (0.021) (0.029) (0.018) (0.043) (0.018) (0.019) (0.031) SMB 0.316*** 0.232*** 0.094* 0.081 0.005 − 0.071 − 0.011 − 0.105 − 0.129*** − 0.218*** (0.084) (0.051) (0.052) (0.063) (0.057) (0.066) (0.061) (0.066) (0.046) (0.056) HML − 0.125** 0.017 0.112*** 0.001 − 0.014 0.058* 0.066 0.167*** − 0.014 0.056 (0.054) (0.035) (0.038) (0.042) (0.058) (0.035) (0.078) (0.038) (0.062) (0.047) WML − 0.041 − 0.040* − 0.010 − 0.030 0.018 0.052*** − 0.029 − 0.018 0.008 0.082*** (0.037) (0.021) (0.021) (0.024) (0.041) (0.020) (0.032) (0.049) (0.016) (0.024) UMS 0.531*** 0.540*** 0.386*** 0.229*** 0.142*** − 0.032 − 0.155** − 0.211*** − 0.234*** − 0.282*** (0.082) (0.042) (0.061) (0.065) (0.052) (0.047) (0.065) (0.046) (0.031) (0.032) Constant − 0.001 − 0.0004 − 0.0002 − 0.0004 0.001 − 0.0002 − 0.001 − 0.001 − 0.001 0.00004 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 289 289 289 289 289 288 289 289 288 289 Adjusted R20.921*** 0.965*** 0.956*** 0.963*** 0.969*** 0.974 0.970*** 0.969*** 0.979*** 0.969*** Social Market 0.986*** 0.972*** 1.097*** 1.071*** 1.090*** 1.019*** 1.021*** 0.954*** 0.995*** 0.972*** (0.025) (0.016) (0.021) (0.025) (0.036) (0.022) (0.031) (0.014) (0.023) (0.022) SMB 0.234*** 0.095** 0.177*** 0.113* − 0.018 0.121** 0.018 − 0.039 − 0.153*** − 0.238*** (0.063) (0.045) (0.059) (0.065) (0.072) (0.061) (0.062) (0.041) (0.056) (0.047) HML − 0.074 − 0.037 − 0.105 0.009 − 0.132 0.088 0.230*** 0.072 0.086** 0.040 (0.054) (0.052) (0.088) (0.072) (0.098) (0.063) (0.079) (0.051) (0.040) (0.063) WML − 0.009 − 0.034 − 0.030 − 0.002 − 0.039 − 0.048 0.048 0.008 0.016 0.041** (0.032) (0.027) (0.032) (0.025) (0.026) (0.034) (0.043) (0.020) (0.025) (0.018) UMS 0.360*** 0.403*** 0.301*** 0.296*** 0.275*** 0.216*** 0.202*** − 0.070 − 0.100 − 0.448*** (0.134) (0.102) (0.113) (0.105) (0.094) (0.081) (0.069) (0.055) (0.065) (0.067) Constant 0.0004 0.001* 0.001 0.001 − 0.0003 0.001 − 0.0001 0.0001 − 0.002*** − 0.0002 (0.001) (0.0005) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 289 289 288 289 289 288 289 289 289 289 Adjusted R20.952*** 0.967*** 0.966*** 0.961*** 0.952*** 0.965*** 0.969*** 0.976 0.978* 0.976*** 50 J.Dobrick et al. not priced, which in our opinion is however not of great concern, since the “standard” factors have themselves been not priced on average for the same time period. Common variation overtime In this section, we describe the results obtained from our time-series regressions with respect to whether the constructed risk factors related to ESG capture common variation in returns over time for the global market. The interpretation of the results refers to those regressions conducted with UMS factors built upon ESG data of t-2. Regressions conducted with ESG factor construction based on data of t-1 for robustness reasons however lead to results congruent with our prior findings.1 The time-series regression outputs can be found in Tables3, 4 and 5, and our Table 3 (continued) Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) ES Market 0.988*** 1.041*** 1.022*** 1.085*** 1.041*** 1.035*** 1.065*** 0.998*** 0.946*** 0.989*** (0.030) (0.025) (0.016) (0.024) (0.023) (0.027) (0.030) (0.020) (0.015) (0.022) SMB 0.180*** 0.132* 0.119*** 0.199*** 0.016 0.016 0.072 0.086 − 0.163*** − 0.261*** (0.053) (0.069) (0.045) (0.065) (0.071) (0.063) (0.062) (0.061) (0.041) (0.045) HML − 0.052 − 0.008 − 0.063 − 0.047 0.035 0.127 − 0.010 0.200** 0.067 0.018 (0.081) (0.078) (0.076) (0.060) (0.097) (0.093) (0.080) (0.084) (0.044) (0.063) WML − 0.018 0.008 − 0.066* 0.014 − 0.019 − 0.045 − 0.058** 0.034 0.041** 0.036** (0.029) (0.025) (0.035) (0.029) (0.038) (0.033) (0.023) (0.024) (0.018) (0.015) UMS 0.454*** 0.468*** 0.315*** 0.281*** 0.259*** 0.224** 0.136 0.036 − 0.181*** − 0.404*** (0.103) (0.072) (0.077) (0.067) (0.081) (0.093) (0.090) (0.069) (0.062) (0.060) Constant 0.001 0.001 0.001 0.001 0.0004 − 0.001 − 0.001 0.0002 − 0.001* − 0.0004 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 289 289 288 289 289 289 288 288 289 289 Adjusted R20.950*** 0.971*** 0.968*** 0.964*** 0.964*** 0.962*** 0.961* 0.971 0.981*** 0.979*** ESG Market 0.956*** 1.084*** 1.051*** 1.016*** 1.087*** 1.024*** 0.936*** 1.024*** 1.053*** 0.939*** (0.036) (0.026) (0.038) (0.022) (0.042) (0.025) (0.018) (0.027) (0.024) (0.024) SMB 0.161*** 0.231*** 0.214*** 0.153*** 0.145* 0.078* 0.089** 0.069 − 0.069 − 0.325*** (0.046) (0.051) (0.071) (0.039) (0.078) (0.043) (0.041) (0.070) (0.052) (0.044) HML − 0.110 − 0.168*** 0.022 − 0.067** − 0.068 0.109** 0.019 0.058 0.069 0.084* (0.075) (0.047) (0.079) (0.033) (0.087) (0.046) (0.058) (0.067) (0.055) (0.046) WML − 0.039 − 0.061*** − 0.032 − 0.012 − 0.057** − 0.058*** − 0.035 0.038 0.062** 0.049*** (0.037) (0.020) (0.032) (0.021) (0.027) (0.017) (0.022) (0.034) (0.024) (0.018) UMS 0.484*** 0.304*** 0.223* 0.065 0.172 0.274*** − 0.001 − 0.110*** − 0.238*** − 0.230*** (0.107) (0.077) (0.135) (0.080) (0.105) (0.039) (0.085) (0.042) (0.072) (0.045) Constant − 0.001 − 0.0003 0.002** 0.001* − 0.001 − 0.001* − 0.001 0.0002 − 0.0003 − 0.0004 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 289 289 289 289 289 288 289 289 288 289 Adjusted R20.927*** 0.967*** 0.956*** 0.961 0.954** 0.973*** 0.965 0.970* 0.979*** 0.980*** This table shows the regression results of value-weighted decile portfolios for five different ESG dimensions. The utilized model is the Carhart four-factor model with an additional UMS factor accounting for each dimension of ESG *, ** and *** denote significance levels on the 10%, 5% and 1% significance levels, respectively. Significance level indications pertaining to the Carhart factors, UMS and constants are obtained from t-tests, adjusted R2 indications are from F-tests for nested models. Newey–West standard errors are reported in parentheses 1 Due to space constraints we do only report the statistics for the Carhart + UMS regressions with ESG data as of t-2.The results of the regressions conducted with ESG data of t-1 as well as the results for the Fama–French + UMS regressions are available on request. 51ESG asrisk factor Table 4 Refinitiv world: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 1.039*** 0.931*** 0.737*** 0.989*** 1.072*** 1.050*** 0.987*** 0.970*** 0.978*** 0.997*** (0.023) (0.058) (0.143) (0.065) (0.029) (0.028) (0.035) (0.023) (0.019) (0.022) SMB − 0.022 0.314** 0.133 0.302** 0.171** 0.155*** 0.141* − 0.056 − 0.195*** − 0.249*** (0.064) (0.123) (0.169) (0.128) (0.077) (0.048) (0.073) (0.052) (0.039) (0.041) HML − 0.051 − 0.008 0.252* 0.030 − 0.041 − 0.021 0.220*** 0.029 0.059 0.094* (0.053) (0.064) (0.148) (0.085) (0.053) (0.057) (0.063) (0.064) (0.037) (0.054) WML − 0.051 − 0.047 0.284*** − 0.042 − 0.004 − 0.020 0.025 0.010 0.075*** 0.010 (0.032) (0.067) (0.088) (0.065) (0.026) (0.026) (0.035) (0.025) (0.021) (0.022) UMS 0.660*** 0.176 0.383** 0.037 0.097 0.111 0.148* − 0.077 − 0.156** − 0.227*** (0.078) (0.162) (0.152) (0.151) (0.084) (0.082) (0.079) (0.056) (0.070) (0.080) Constant 0.001 0.0003 0.001 0.002* 0.00002 − 0.001 − 0.002** 0.0003 − 0.001* 0.0001 (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 401 212 280 267 284 288 288 288 288 289 Adjusted R20.965*** 0.900* 0.682** 0.912 0.963 0.962 0.966** 0.971 0.973*** 0.976*** Governance Market 0.933*** 1.068*** 1.089*** 1.134*** 1.034*** 1.003*** 1.010*** 0.967*** 0.989*** 0.935*** (0.032) (0.024) (0.015) (0.042) (0.025) (0.021) (0.024) (0.021) (0.034) (0.020) SMB 0.268*** 0.181*** 0.074 0.025 − 0.066 − 0.062 − 0.026 − 0.071 − 0.224** − 0.118*** (0.045) (0.044) (0.061) (0.037) (0.042) (0.060) (0.062) (0.055) (0.099) (0.037) HML 0.047 0.042 − 0.025 − 0.062 − 0.003 0.120*** 0.094* 0.064 − 0.014 0.129*** (0.048) (0.036) (0.039) (0.044) (0.034) (0.046) (0.053) (0.065) (0.080) (0.037) WML − 0.038 − 0.016 − 0.013 − 0.047** − 0.001 0.013 0.026 − 0.023 − 0.042 0.109*** (0.030) (0.022) (0.016) (0.024) (0.022) (0.032) (0.031) (0.058) (0.029) (0.025) UMS 0.653*** 0.563*** 0.412*** 0.193*** 0.042 − 0.111* − 0.169*** − 0.257*** − 0.253*** − 0.231*** (0.058) (0.045) (0.069) (0.041) (0.059) (0.057) (0.033) (0.038) (0.049) (0.032) Constant − 0.0003 − 0.001 0.0002 − 0.0002 0.0001 0.0002 − 0.001 − 0.001 − 0.0002 − 0.0005 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 291 286 288 288 288 288 288 289 288 289 Adjusted R20.944*** 0.971*** 0.965*** 0.975*** 0.974 0.972*** 0.974*** 0.966*** 0.959*** 0.977*** Social Market 0.907*** 1.013*** 1.104*** 1.077*** 1.032*** 1.047*** 1.050*** 0.981*** 1.022*** 0.958*** (0.038) (0.018) (0.034) (0.025) (0.023) (0.021) (0.030) (0.033) (0.018) (0.016) SMB 0.207*** 0.089 0.159*** 0.149*** 0.080 0.067 0.048 − 0.030 − 0.006 − 0.320*** (0.062) (0.057) (0.054) (0.050) (0.055) (0.043) (0.055) (0.048) (0.045) (0.031) HML − 0.016 − 0.021 − 0.019 0.085* 0.166*** 0.218*** 0.074 0.002 0.064* 0.018 (0.053) (0.035) (0.058) (0.050) (0.043) (0.044) (0.065) (0.076) (0.033) (0.046) WML − 0.050* − 0.015 − 0.109*** − 0.035 − 0.073*** − 0.029* 0.023 − 0.033 − 0.035* 0.111*** (0.027) (0.026) (0.037) (0.025) (0.023) (0.015) (0.019) (0.024) (0.021) (0.025) UMS 0.477*** 0.548*** 0.525*** 0.257*** 0.288*** 0.308*** − 0.053 − 0.024 − 0.281*** − 0.210*** (0.085) (0.064) (0.103) (0.090) (0.080) (0.064) (0.057) (0.058) (0.060) (0.074) Constant − 0.001 0.001* 0.0003 0.001 0.0005 − 0.00004 − 0.0001 − 0.0004 − 0.001 − 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 289 288 288 289 288 288 288 289 288 289 Adjusted R20.932*** 0.967*** 0.963*** 0.971*** 0.969*** 0.978*** 0.979 0.964 0.981*** 0.979*** ES Market 0.954*** 1.043*** 1.033*** 1.018*** 1.087*** 1.070*** 0.976*** 1.009*** 1.023*** 0.967*** (0.039) (0.036) (0.020) (0.027) (0.021) (0.020) (0.027) (0.027) (0.013) (0.019) 58 J.Dobrick et al. This table shows the regression results of value-weighted decile portfolios for five different ESG dimensions. The utilized model is the Carhart four-factor model with an additional UMS factor accounting for each dimension of ESG. *, ** and *** denote significance levels on the 10%, 5% and 1% significance levels, respectively. Significance level indications pertaining to the Carhart factors, UMS and constants are obtained from t-tests, and adjusted R2 indications are from F-tests for nested models. Newey–West standard errors are reported in parentheses. Table 8 (continued) Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) SMB 0.196*** 0.165*** 0.137 0.151** 0.155* 0.180*** − 0.074 0.113** − 0.033 − 0.221*** (0.069) (0.054) (0.084) (0.063) (0.087) (0.046) (0.048) (0.050) (0.049) (0.035) HML − 0.045 − 0.102 0.101 − 0.035 0.072 0.128*** − 0.263*** 0.083 0.198*** − 0.006 (0.095) (0.082) (0.112) (0.093) (0.058) (0.039) (0.079) (0.064) (0.046) (0.048) WML − 0.028 − 0.225*** 0.059 0.035 0.118* 0.107*** − 0.152** 0.089* 0.129*** − 0.075 (0.083) (0.085) (0.066) (0.057) (0.061) (0.035) (0.060) (0.049) (0.041) (0.046) UMS 0.747*** 0.753*** 0.495*** 0.159** 0.207* 0.069 − 0.091 − 0.044 − 0.240*** − 0.195*** (0.109) (0.134) (0.095) (0.080) (0.113) (0.071) (0.100) (0.091) (0.077) (0.039) Constant − 0.001 − 0.001 0.001 0.00000 0.001 0.001 − 0.001 0.001 0.0005 − 0.0002 (0.002) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 99 99 98 99 99 98 98 99 99 99 Adjusted R20.881*** 0.908*** 0.922 0.916 0.911* 0.946 0.892 0.923 0.935*** 0.957*** ESG Market 1.181*** 1.051*** 1.131*** 1.138*** 1.072*** 1.061*** 1.075*** 1.049*** 0.989*** 0.948*** (0.062) (0.032) (0.068) (0.040) (0.051) (0.035) (0.050) (0.031) (0.034) (0.026) SMB 0.095 0.233** 0.239** 0.131* 0.089 0.110** 0.005 0.100** 0.009 − 0.266*** (0.075) (0.092) (0.093) (0.077) (0.067) (0.053) (0.052) (0.042) (0.040) (0.047) HML − 0.165** − 0.162** 0.144* − 0.100 0.143** 0.053 − 0.145** 0.129** 0.190*** − 0.086** (0.077) (0.076) (0.087) (0.087) (0.057) (0.052) (0.071) (0.053) (0.036) (0.040) WML − 0.232** − 0.047 0.085 − 0.065 0.187*** − 0.025 − 0.063 0.133*** 0.102** − 0.086** (0.099) (0.048) (0.063) (0.069) (0.046) (0.041) (0.060) (0.034) (0.047) (0.042) UMS 0.833*** 0.576*** 0.378*** 0.329** 0.107 0.142*** − 0.108 − 0.131 − 0.157*** − 0.226*** (0.116) (0.089) (0.102) (0.135) (0.077) (0.055) (0.101) (0.089) (0.061) (0.049) Constant − 0.001 0.002 0.00001 − 0.0003 0.002 0.0004 − 0.001 − 0.00003 0.0005 0.0001 (0.002) (0.002) (0.001) (0.001) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) Observations 99 99 99 99 98 98 99 99 98 99 Adjusted R20.880*** 0.893*** 0.918*** 0.913*** 0.896 0.930 0.910 0.928 0.951** 0.956*** This table shows the regression results of value-weighted decile portfolios for five different ESG dimensions. The utilized model is the Carhart four-factor model with an additional UMS factor accounting for each dimension of ESG *, ** and *** denote significance levels on the 10%, 5% and 1% significance levels, respectively Significance level indications pertaining to the Carhart factors, UMS and constants are obtained from t-tests, and adjusted R2 indications are from F-tests for nested models. Newey–West standard errors are reported in parentheses 59ESG asrisk factor Table 9 Refinitiv North America: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 1.110*** 0.672*** 1.108*** 1.049*** 1.129*** 1.208*** 1.100*** 0.985*** 0.942*** 1.009*** (0.028) (0.214) (0.034) (0.031) (0.067) (0.051) (0.034) (0.037) (0.027) (0.022) SMB 0.195*** 0.419*** 0.235*** 0.281*** 0.206** 0.073 0.080 0.118** − 0.088* − 0.196*** (0.042) (0.135) (0.051) (0.051) (0.094) (0.067) (0.055) (0.056) (0.049) (0.032) HML − 0.112** 0.131 0.138* 0.034 0.052 0.270*** 0.009 0.126*** 0.114*** − 0.089** (0.053) (0.128) (0.078) (0.054) (0.109) (0.057) (0.073) (0.046) (0.030) (0.038) WML − 0.081 0.003 − 0.087 0.119*** 0.105* 0.076 − 0.033 − 0.037 0.050 − 0.017 (0.057) (0.182) (0.072) (0.046) (0.059) (0.055) (0.048) (0.056) (0.045) (0.032) UMS 0.691*** − 0.248 0.478*** 0.127 0.0003 0.164** 0.237** − 0.071 − 0.020 − 0.301*** (0.092) (0.393) (0.137) (0.132) (0.086) (0.068) (0.097) (0.051) (0.089) (0.053) Constant 0.001 − 0.004 − 0.002 0.001 0.002 0.001 − 0.00001 − 0.001 − 0.0004 0.0001 (0.001) (0.003) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 187 21 93 95 99 98 99 98 99 99 Adjusted R20.951*** 0.574 0.921*** 0.931 0.916 0.924 0.916** 0.944 0.929 0.961*** Governance Market 1.055*** 1.018*** 1.091*** 1.048*** 1.021*** 1.151*** 0.985*** 1.043*** 1.076*** 0.925*** (0.048) (0.052) (0.038) (0.037) (0.043) (0.043) (0.025) (0.033) (0.026) (0.037) SMB 0.185*** 0.050 0.161*** − 0.005 0.065 0.024 − 0.003 − 0.098* − 0.047 − 0.048 (0.067) (0.067) (0.048) (0.069) (0.066) (0.062) (0.049) (0.054) (0.047) (0.058) HML − 0.055 − 0.098 0.036 0.097* 0.041 0.147 0.101** − 0.088 0.024 − 0.045 (0.073) (0.067) (0.043) (0.053) (0.053) (0.097) (0.041) (0.057) (0.050) (0.062) WML − 0.139 − 0.003 0.025 0.039 0.002 − 0.024 0.047 0.066 0.001 − 0.046 (0.094) (0.091) (0.051) (0.058) (0.063) (0.074) (0.050) (0.048) (0.043) (0.052) UMS 0.620*** 0.399*** 0.289*** 0.111 − 0.104 − 0.022 − 0.037 − 0.173*** − 0.127*** − 0.302*** (0.133) (0.112) (0.072) (0.072) (0.075) (0.093) (0.059) (0.061) (0.043) (0.062) Constant 0.0001 − 0.001 − 0.0001 0.001 0.002** − 0.002* 0.001 − 0.001 0.0003 0.001 (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 104 101 101 102 102 101 102 101 102 102 Adjusted R20.896*** 0.900*** 0.955*** 0.935 0.917 0.928 0.926 0.921** 0.949** 0.924*** Social Market 1.046*** 1.045*** 1.168*** 1.157*** 1.074*** 1.170*** 1.043*** 1.056*** 0.997*** 0.968*** (0.031) (0.036) (0.039) (0.054) (0.032) (0.036) (0.039) (0.041) (0.046) (0.017) SMB 0.101 0.155*** 0.247*** 0.249*** 0.276*** 0.198*** 0.016 − 0.007 − 0.042 − 0.222*** (0.083) (0.051) (0.056) (0.070) (0.068) (0.067) (0.049) (0.061) (0.066) (0.037) HML − 0.067 − 0.093 0.132** 0.084 0.227*** 0.279*** 0.127* 0.008 − 0.086 − 0.012 (0.062) (0.064) (0.059) (0.084) (0.061) (0.085) (0.073) (0.054) (0.063) (0.037) WML − 0.025 − 0.028 − 0.004 0.053 0.004 0.007 − 0.073 0.024 0.045 0.010 (0.082) (0.063) (0.052) (0.088) (0.093) (0.076) (0.045) (0.045) (0.050) (0.050) UMS 0.447*** 0.439*** 0.390*** 0.254*** 0.192 0.357*** 0.109 − 0.085 − 0.289*** − 0.201*** (0.134) (0.074) (0.102) (0.078) (0.117) (0.085) (0.073) (0.070) (0.057) (0.068) Constant − 0.0004 0.001 0.0003 0.002*** 0.0002 0.0001 0.002 0.00001 0.0002 − 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 102 102 101 102 102 101 102 101 102 102 Adjusted R20.887*** 0.916*** 0.938*** 0.937*** 0.913* 0.930*** 0.946 0.921 0.927*** 0.952*** ES Market 1.087*** 1.308*** 1.125*** 1.000*** 1.071*** 1.142*** 1.124*** 1.140*** 1.047*** 0.971*** (0.034) (0.064) (0.040) (0.034) (0.040) (0.063) (0.045) (0.027) (0.028) (0.017) 60 J.Dobrick et al. Table 9 (continued) Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) SMB 0.167* 0.274*** 0.193*** 0.202*** 0.262*** 0.149* 0.155*** 0.092* − 0.041 − 0.216*** (0.091) (0.068) (0.065) (0.049) (0.061) (0.090) (0.057) (0.049) (0.043) (0.032) HML − 0.203** 0.025 0.145* − 0.060 0.183*** 0.153 0.231*** 0.072* 0.100* − 0.111*** (0.084) (0.076) (0.079) (0.047) (0.059) (0.107) (0.055) (0.037) (0.053) (0.043) WML − 0.089 − 0.137 0.036 − 0.114** 0.112** − 0.041 0.139*** − 0.034 − 0.023 0.018 (0.069) (0.104) (0.069) (0.047) (0.053) (0.061) (0.052) (0.040) (0.048) (0.036) UMS 0.514*** 0.794*** 0.536*** 0.147** 0.194* 0.027 0.110** 0.123* − 0.011 − 0.292*** (0.142) (0.171) (0.111) (0.074) (0.112) (0.073) (0.054) (0.072) (0.076) (0.050) Constant 0.001 − 0.003* 0.001 − 0.001 0.001 0.0002 0.002* − 0.001 − 0.0001 − 0.0001 (0.001) (0.001) (0.002) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) Observations 100 102 98 98 97 99 98 99 98 99 Adjusted s20.890*** 0.895*** 0.907*** 0.939* 0.938** 0.925 0.933 0.955 0.950 0.964*** ESG Market 1.265*** 1.168*** 0.934*** 1.044*** 1.078*** 1.106*** 1.191*** 1.162*** 1.017*** 0.974*** (0.049) (0.041) (0.035) (0.046) (0.069) (0.040) (0.046) (0.031) (0.029) (0.024) SMB 0.147 0.222*** 0.328*** 0.232*** 0.201*** 0.172** 0.152** 0.068 − 0.018 − 0.231*** (0.097) (0.067) (0.067) (0.059) (0.063) (0.082) (0.073) (0.042) (0.043) (0.036) HML − 0.236** 0.012 0.135*** 0.017 0.161** 0.137 0.188*** 0.108* 0.138*** − 0.134*** (0.104) (0.069) (0.051) (0.073) (0.073) (0.085) (0.040) (0.056) (0.047) (0.046) WML − 0.154* 0.006 − 0.052 0.083 0.099 − 0.003 0.068 0.069 − 0.029 − 0.007 (0.092) (0.047) (0.044) (0.064) (0.067) (0.044) (0.061) (0.049) (0.043) (0.042) UMS 0.692*** 0.360*** 0.611*** 0.056 0.091 − 0.016 0.201** 0.039 0.006 − 0.290*** (0.165) (0.060) (0.079) (0.083) (0.106) (0.070) (0.089) (0.069) (0.083) (0.049) Constant − 0.001 0.001 0.00001 0.001 0.001 0.001 0.001 − 0.001* − 0.0004 0.0002 (0.002) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 99 101 99 97 98 99 99 98 99 99 Adjusted R20.882*** 0.953*** 0.887*** 0.927 0.926 0.937 0.924** 0.951 0.946 0.960*** This table shows the regression results of value-weighted decile portfolios for five different ESG dimensions. The utilized model is the Carhart four-factor model with an additional UMS factor accounting for each dimension of ESG *, ** and *** denote significance levels on the 10%, 5% and 1% significance levels, respectively. Significance level indications pertaining to the Carhart factors, UMS and constants are obtained from t-tests, and adjusted R2 indications are from F-tests for nested models. Newey–West standard errors are reported in parentheses 61ESG asrisk factor Table 10 Vigeo Eiris North America: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 1.090*** 1.085*** 0.957*** 1.031*** 1.139*** 1.040*** 1.011*** 1.105*** 1.016*** 1.035*** (0.070) (0.040) (0.042) (0.035) (0.054) (0.031) (0.040) (0.064) (0.030) (0.019) SMB 0.127 0.109 0.157** 0.150** 0.133* 0.033 0.128** − 0.058 − 0.140** − 0.282*** (0.086) (0.066) (0.078) (0.059) (0.073) (0.066) (0.063) (0.081) (0.063) (0.028) HML − 0.223** − 0.089 − 0.038 0.092 0.088 − 0.011 0.172* 0.134 0.036 0.020 (0.093) (0.090) (0.061) (0.077) (0.064) (0.093) (0.093) (0.125) (0.075) (0.034) WML 0.007 0.004 0.049 − 0.034 0.106* − 0.041 − 0.002 0.023 0.101* − 0.114*** (0.101) (0.070) (0.048) (0.037) (0.056) (0.047) (0.072) (0.071) (0.052) (0.028) UMS 0.149** − 0.039 0.044 0.056 0.007 0.064* − 0.160*** − 0.101* 0.037 − 0.002 (0.070) (0.047) (0.035) (0.034) (0.040) (0.037) (0.058) (0.055) (0.031) (0.026) Constant − 0.001 − 0.0002 0.002 0.002 0.0002 0.001 − 0.004** − 0.003** 0.001 0.001 (0.002) (0.001) (0.001) (0.001) (0.002) (0.001) (0.002) (0.001) (0.001) (0.001) Observations 64 57 61 62 54 58 56 58 56 56 Adjusted R20.822** 0.916 0.904 0.939 0.887 0.902 0.893*** 0.906** 0.902 0.951 Governance Market 1.048*** 1.029*** 1.116*** 1.062*** 1.011*** 1.087*** 1.039*** 0.992*** 1.002*** 1.033*** (0.100) (0.035) (0.054) (0.034) (0.051) (0.042) (0.049) (0.039) (0.053) (0.037) SMB 0.106 0.023 0.090 − 0.033 − 0.051 − 0.048 − 0.172** − 0.249*** 0.047 0.038 (0.083) (0.080) (0.085) (0.052) (0.072) (0.055) (0.070) (0.067) (0.081) (0.060) HML − 0.344*** − 0.046 0.165** − 0.059 − 0.053 − 0.056 0.099 0.021 0.348*** 0.085* (0.100) (0.064) (0.068) (0.071) (0.089) (0.057) (0.083) (0.070) (0.075) (0.050) WML − 0.176 − 0.070 0.092* − 0.008 − 0.036 0.066 0.047 − 0.016 0.098 − 0.033 (0.121) (0.051) (0.053) (0.049) (0.065) (0.064) (0.048) (0.069) (0.079) (0.045) UMS 0.078 0.008 0.139 − 0.011 − 0.009 − 0.053 0.048 − 0.107** − 0.055 − 0.097** (0.094) (0.049) (0.090) (0.045) (0.062) (0.060) (0.072) (0.049) (0.048) (0.042) Constant − 0.002 0.002 − 0.002** 0.002** − 0.002 − 0.001 0.001 0.002 − 0.002 − 0.001 (0.002) (0.002) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 63 61 56 56 59 58 57 57 59 56 Adjusted R20.824 0.902 0.915** 0.919 0.838 0.908 0.916 0.882* 0.905 0.933* Social Market 0.958*** 1.141*** 1.075*** 1.103*** 1.048*** 1.117*** 1.061*** 1.023*** 1.056*** 0.933*** (0.036) (0.042) (0.054) (0.042) (0.039) (0.030) (0.037) (0.029) (0.028) (0.028) SMB 0.229*** 0.120* − 0.073 0.173** − 0.007 0.109** 0.0001 − 0.067 − 0.097** − 0.307*** (0.070) (0.069) (0.087) (0.087) (0.063) (0.055) (0.066) (0.048) (0.045) (0.046) HML 0.056 − 0.165** − 0.362*** − 0.109** 0.031 0.225*** 0.023 0.202*** 0.119*** 0.052 (0.075) (0.067) (0.097) (0.053) (0.076) (0.079) (0.060) (0.054) (0.030) (0.036) WML 0.108 0.079 0.012 − 0.050 − 0.062 0.047 0.059 0.017 0.014 − 0.096 (0.069) (0.109) (0.087) (0.066) (0.042) (0.058) (0.055) (0.048) (0.055) (0.059) UMS 0.014 0.073** 0.073** 0.037 0.005 0.014 0.031* 0.016 − 0.082*** − 0.046** (0.029) (0.034) (0.033) (0.029) (0.037) (0.025) (0.019) (0.024) (0.025) (0.023) Constant − 0.001 0.003 0.002 0.002 − 0.002 0.001 0.0001 0.001 − 0.002** − 0.002* (0.001) (0.002) (0.002) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 62 59 56 59 59 57 58 59 56 57 Adjusted R20.880 0.830 0.798 0.886 0.888 0.937 0.927 0.936 0.948*** 0.929* ES Market 1.109*** 1.092*** 0.939*** 1.063*** 1.046*** 1.068*** 1.140*** 1.086*** 1.009*** 0.990*** (0.048) (0.059) (0.048) (0.041) (0.053) (0.030) (0.077) (0.038) (0.023) (0.017) 62 J.Dobrick et al. Table 10 (continued) Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) SMB 0.094 0.248*** 0.116 0.103* 0.064 0.115** 0.076 − 0.102* − 0.084* − 0.311*** (0.100) (0.066) (0.071) (0.060) (0.077) (0.052) (0.091) (0.061) (0.051) (0.037) HML − 0.132 − 0.146* − 0.058 0.095 − 0.120 0.215*** − 0.026 0.028 0.119*** 0.071** (0.112) (0.081) (0.072) (0.071) (0.105) (0.052) (0.109) (0.073) (0.045) (0.029) WML 0.130 − 0.120 0.050 0.014 0.011 − 0.003 0.129 − 0.062 0.061 − 0.103*** (0.096) (0.087) (0.068) (0.058) (0.055) (0.059) (0.080) (0.068) (0.039) (0.029) UMS 0.060 0.075 0.025 − 0.001 0.022 0.008 − 0.061 − 0.001 0.024 − 0.061** (0.058) (0.048) (0.039) (0.029) (0.035) (0.035) (0.042) (0.050) (0.044) (0.025) Constant − 0.001 − 0.001 0.001 0.003*** 0.0003 − 0.0003 − 0.003* − 0.0002 − 0.0004 − 0.0003 (0.002) (0.001) (0.002) (0.001) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) Observations 63 69 55 62 63 56 58 52 56 49 Adjusted R20.829 0.892 0.879 0.902 0.881 0.945 0.869 0.916 0.934 0.957** ESG Market 1.098*** 0.984*** 0.997*** 1.038*** 1.049*** 1.090*** 1.124*** 1.086*** 1.016*** 0.988*** (0.056) (0.033) (0.050) (0.040) (0.054) (0.035) (0.055) (0.029) (0.031) (0.021) SMB 0.117 0.141** 0.160** 0.131** 0.024 0.003 0.159* − 0.042 − 0.161*** − 0.287*** (0.080) (0.062) (0.074) (0.061) (0.092) (0.061) (0.093) (0.056) (0.045) (0.041) HML − 0.345*** − 0.044 − 0.049 0.148** − 0.144 0.049 0.092 − 0.022 0.206*** 0.061** (0.099) (0.063) (0.057) (0.068) (0.126) (0.096) (0.096) (0.049) (0.049) (0.030) WML − 0.084 0.100* − 0.044 0.101* − 0.014 0.017 0.011 0.029 0.040 − 0.098** (0.107) (0.054) (0.051) (0.053) (0.068) (0.056) (0.077) (0.059) (0.060) (0.039) UMS 0.048 − 0.011 − 0.007 0.091** 0.018 − 0.011 0.098 − 0.110*** 0.008 − 0.057** (0.065) (0.035) (0.036) (0.036) (0.038) (0.036) (0.077) (0.032) (0.037) (0.027) Constant − 0.003 0.001 0.002* 0.001 − 0.00001 − 0.002 0.001 0.002 − 0.001 − 0.001 (0.002) (0.001) (0.001) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) Observations 70 63 56 56 58 56 55 56 58 54 Adjusted R20.825 0.906 0.912 0.908** 0.841 0.897 0.885* 0.921** 0.936 0.946** 63ESG asrisk factor Table 11 ASSET4 Europe: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 1.034*** 0.986*** 1.113*** 1.038*** 1.052*** 1.053*** 1.030*** 0.969*** 0.992*** 1.027*** (0.022) (0.031) (0.030) (0.023) (0.033) (0.016) (0.018) (0.021) (0.017) (0.013) SMB 0.245*** 0.533*** 0.544*** 0.292*** 0.148** − 0.026 − 0.112* − 0.090 − 0.108 − 0.137*** (0.062) (0.092) (0.131) (0.059) (0.068) (0.080) (0.062) (0.077) (0.072) (0.043) HML 0.076 − 0.088 − 0.075 − 0.213*** − 0.038 − 0.204*** 0.082 0.008 0.155** 0.061 (0.070) (0.059) (0.095) (0.050) (0.095) (0.059) (0.059) (0.051) (0.068) (0.047) WML − 0.033 − 0.002 − 0.0002 − 0.068** − 0.114 − 0.126*** − 0.031 0.061* 0.082** 0.009 (0.025) (0.040) (0.036) (0.031) (0.080) (0.048) (0.034) (0.032) (0.039) (0.019) UMS 0.180** 0.325*** 0.247*** − 0.045 0.257*** 0.248*** 0.017 − 0.113* − 0.096* − 0.213*** (0.072) (0.067) (0.072) (0.075) (0.076) (0.063) (0.070) (0.063) (0.058) (0.044) Constant 0.0003 0.0004 0.001 0.0003 0.001 0.001* − 0.001 0.001 − 0.0002 0.0004 (0.001) (0.001) (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 80 79 79 79 80 79 79 80 80 79 Adjusted R2 0.960** 0.952*** 0.955*** 0.964 0.946*** 0.964*** 0.972 0.970** 0.977* 0.983*** Governance Market 1.048*** 1.041*** 1.044*** 1.027*** 1.036*** 1.072*** 1.059*** 0.993*** 1.054*** 0.949*** (0.016) (0.020) (0.024) (0.019) (0.019) (0.029) (0.027) (0.013) (0.036) (0.015) SMB 0.460*** 0.040 0.075 − 0.004 − 0.003 0.025 − 0.104* − 0.102*** − 0.015 − 0.143*** (0.048) (0.051) (0.052) (0.047) (0.092) (0.074) (0.053) (0.038) (0.075) (0.050) HML − 0.127*** − 0.030 − 0.054 0.113* 0.008 0.073 0.016 0.065 0.013 0.037 (0.033) (0.037) (0.056) (0.059) (0.075) (0.060) (0.067) (0.051) (0.052) (0.037) WML − 0.047 − 0.010 0.034 0.119*** 0.027 − 0.004 0.041* 0.009 − 0.095** 0.0001 (0.035) (0.029) (0.028) (0.022) (0.025) (0.057) (0.024) (0.023) (0.039) (0.019) UMS 0.400*** 0.277*** 0.494*** 0.084 0.131* 0.061 − 0.143*** − 0.146*** − 0.097* − 0.323*** (0.052) (0.049) (0.052) (0.055) (0.071) (0.117) (0.048) (0.042) (0.055) (0.072) Constant − 0.0004 0.001 0.002* 0.0001 − 0.0002 0.0003 − 0.002*** 0.001* 0.0001 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 80 80 79 79 80 79 79 80 79 80 Adjusted R2 0.971*** 0.969*** 0.969*** 0.969* 0.969** 0.959 0.963** 0.978*** 0.964 0.974*** Social Market 1.084*** 1.015*** 1.070*** 0.988*** 1.034*** 0.999*** 1.022*** 1.029*** 1.013*** 1.001*** (0.038) (0.029) (0.030) (0.022) (0.044) (0.020) (0.023) (0.021) (0.017) (0.020) SMB 0.405*** 0.351*** 0.432*** 0.350*** 0.152 0.159* 0.077 0.003 − 0.166*** − 0.295*** (0.071) (0.062) (0.112) (0.076) (0.097) (0.085) (0.075) (0.083) (0.054) (0.054) HML − 0.039 − 0.042 − 0.041 − 0.078 − 0.016 0.050 0.026 0.069 − 0.057* 0.054 (0.068) (0.057) (0.091) (0.064) (0.109) (0.081) (0.049) (0.059) (0.033) (0.074) WML − 0.085 0.040 − 0.062* 0.037 − 0.007 0.077*** 0.013 0.074*** − 0.044*** − 0.008 (0.075) (0.037) (0.037) (0.033) (0.056) (0.029) (0.032) (0.019) (0.017) (0.027) UMS 0.327*** 0.307*** 0.307*** 0.220*** 0.029 0.087 − 0.063 − 0.060 − 0.109** − 0.209*** (0.080) (0.085) (0.101) (0.072) (0.094) (0.090) (0.064) (0.056) (0.049) (0.077) Constant 0.001 − 0.00004 − 0.0003 0.002 − 0.001 0.002* − 0.0003 0.0003 − 0.0004 0.001 (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 80 79 79 79 79 79 79 79 79 80 64 J.Dobrick et al. Table 11 (continued) Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Adjusted R20.950*** 0.952*** 0.953*** 0.955*** 0.927 0.962 0.960 0.975 0.983** 0.973*** ES Market 1.033*** 1.077*** 1.079*** 0.998*** 1.053*** 1.009*** 1.008*** 0.998*** 1.047*** 0.986*** (0.022) (0.026) (0.036) (0.021) (0.032) (0.033) (0.022) (0.021) (0.016) (0.014) SMB 0.376*** 0.407*** 0.391*** 0.408*** 0.254*** 0.086 0.117 − 0.083 − 0.091** − 0.269*** (0.051) (0.066) (0.079) (0.098) (0.082) (0.067) (0.083) (0.054) (0.040) (0.036) HML − 0.029 − 0.069 0.012 − 0.160** − 0.177*** − 0.057 0.064 0.118** − 0.051 0.087* (0.077) (0.070) (0.053) (0.069) (0.065) (0.089) (0.085) (0.050) (0.034) (0.048) WML − 0.026 − 0.080** − 0.003 0.027 − 0.055 − 0.046 − 0.027 0.103*** − 0.038** 0.027 (0.037) (0.039) (0.034) (0.044) (0.035) (0.047) (0.032) (0.021) (0.015) (0.028) UMS 0.331*** 0.337*** 0.089 0.121 0.108* 0.023 0.136 − 0.123** − 0.178*** − 0.099** (0.083) (0.074) (0.083) (0.118) (0.061) (0.068) (0.092) (0.059) (0.041) (0.042) Constant 0.001 − 0.0003 − 0.0001 − 0.0001 0.0001 0.001 0.001 − 0.001 0.0004 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 80 80 79 79 80 79 79 79 79 79 Adjusted R20.950*** 0.956*** 0.964 0.937 0.961 0.942 0.963** 0.971** 0.985*** 0.981** ESG Market 1.052*** 1.059*** 1.017*** 1.041*** 1.102*** 1.050*** 1.020*** 1.047*** 1.037*** 0.954*** (0.022) (0.023) (0.032) (0.020) (0.042) (0.027) (0.037) (0.015) (0.012) (0.012) SMB 0.407*** 0.376*** 0.293*** 0.321*** 0.216*** 0.153** − 0.025 − 0.057 − 0.114*** − 0.217*** (0.057) (0.053) (0.061) (0.061) (0.081) (0.069) (0.078) (0.086) (0.040) (0.039) HML − 0.094 0.027 − 0.030 − 0.106** − 0.274*** − 0.099 0.162** 0.121* 0.062** 0.032 (0.068) (0.050) (0.061) (0.053) (0.072) (0.063) (0.068) (0.064) (0.028) (0.046) WML − 0.043 − 0.023 − 0.017 − 0.015 − 0.037 − 0.036 0.055* 0.105*** 0.001 − 0.021 (0.052) (0.033) (0.046) (0.038) (0.065) (0.028) (0.028) (0.022) (0.033) (0.017) UMS 0.380*** 0.172* 0.426*** 0.023 − 0.007 0.084 0.033 − 0.191*** − 0.024 − 0.160*** (0.075) (0.094) (0.092) (0.075) (0.108) (0.072) (0.055) (0.071) (0.054) (0.048) Constant 0.0004 − 0.00000 − 0.002 0.002 0.0002 0.0003 0.002 0.001 − 0.001* 0.0003 (0.001) (0.001) (0.002) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 80 79 79 79 79 79 79 80 79 79 Adjusted R20.948*** 0.961** 0.938*** 0.956 0.942 0.961 0.965 0.964*** 0.986 0.981*** This table shows the regression results of value-weighted decile portfolios for five different ESG dimensions. The utilized model is the Carhart four-factor model with an additional UMS factor accounting for each dimension of ESG *, ** and *** denote significance levels on the 10%, 5% and 1% significance levels, respectively. Significance level indications pertaining to the Carhart factors, UMS and constants are obtained from t-tests, and adjusted R2 indications are from F-tests for nested models. Newey–West standard errors are reported in parentheses 65ESG asrisk factor Table 12 Refinitiv Europe: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 0.998*** 1.048*** 1.184*** 1.054*** 1.025*** 1.028*** 1.026*** 1.030*** 0.963*** 1.014*** (0.026) (0.022) (0.030) (0.035) (0.026) (0.025) (0.018) (0.016) (0.019) (0.022) SMB 0.213*** 0.627*** 0.384*** 0.524*** 0.213*** 0.030 − 0.111* − 0.104* − 0.074* − 0.203*** (0.078) (0.087) (0.065) (0.157) (0.075) (0.085) (0.060) (0.053) (0.039) (0.045) HML 0.083 − 0.106 − 0.209*** − 0.042 − 0.212*** − 0.060 0.102*** − 0.069 − 0.019 0.193*** (0.061) (0.067) (0.075) (0.105) (0.065) (0.041) (0.038) (0.048) (0.041) (0.050) WML − 0.028 − 0.018 − 0.072* − 0.028 − 0.175** − 0.026 0.060*** − 0.003 0.073*** − 0.003 (0.032) (0.029) (0.038) (0.039) (0.088) (0.025) (0.017) (0.040) (0.024) (0.041) UMS 0.295*** 0.262*** 0.340*** 0.076 0.149 0.168*** 0.144*** − 0.101** − 0.224*** − 0.131** (0.098) (0.091) (0.091) (0.147) (0.092) (0.062) (0.045) (0.042) (0.031) (0.053) Constant 0.001 − 0.0002 − 0.001 − 0.001 0.001 − 0.00002 0.001 − 0.0002 − 0.001 0.001 (0.002) (0.001) (0.001) (0.001) (0.002) (0.0004) (0.001) (0.001) (0.001) (0.001) Observations 83 84 72 78 79 79 79 79 79 80 Adjusted R20.947*** 0.959*** 0.945*** 0.930 0.946* 0.966** 0.973** 0.972* 0.977*** 0.980** Governance Market 1.062*** 1.021*** 1.007*** 1.059*** 1.079*** 1.020*** 1.003*** 1.033*** 1.014*** 0.964*** (0.018) (0.031) (0.016) (0.031) (0.018) (0.025) (0.026) (0.051) (0.058) (0.027) SMB 0.198*** 0.005 0.007 − 0.129** 0.033 − 0.192*** 0.019 − 0.067 − 0.004 − 0.099 (0.048) (0.064) (0.050) (0.062) (0.055) (0.064) (0.056) (0.085) (0.108) (0.070) HML − 0.113** − 0.031 0.094** 0.095* − 0.139 0.064 0.134 0.035 − 0.007 0.107*** (0.051) (0.046) (0.043) (0.049) (0.089) (0.060) (0.086) (0.057) (0.057) (0.041) WML − 0.019 0.038 0.011 0.042* − 0.094*** − 0.043* 0.037 0.058* − 0.063** 0.028 (0.023) (0.034) (0.025) (0.024) (0.035) (0.024) (0.030) (0.035) (0.026) (0.022) UMS 0.490*** 0.432*** 0.283*** 0.028 0.085 0.102* − 0.139* − 0.178*** − 0.156*** − 0.385*** (0.078) (0.085) (0.060) (0.085) (0.080) (0.059) (0.083) (0.048) (0.053) (0.063) Constant 0.002* − 0.0004 − 0.001 0.0005 0.0004 0.001 − 0.0002 0.001 0.001 − 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 83 83 83 83 83 83 83 83 83 83 Adjusted R20.964*** 0.964*** 0.965*** 0.956 0.955 0.967 0.959** 0.968*** 0.961** 0.964*** Social Market 1.050*** 0.992*** 1.033*** 1.089*** 1.063*** 1.020*** 1.053*** 1.029*** 1.014*** 0.973*** (0.025) (0.034) (0.025) (0.026) (0.021) (0.021) (0.020) (0.018) (0.015) (0.009) SMB 0.420*** 0.329*** 0.076 0.206*** 0.115** 0.079 0.011 0.063 − 0.097** − 0.219*** (0.067) (0.068) (0.060) (0.078) (0.046) (0.110) (0.067) (0.065) (0.040) (0.028) HML 0.073 0.057 − 0.050 0.008 0.196*** 0.322*** 0.233*** 0.159*** 0.017 − 0.171*** (0.059) (0.057) (0.046) (0.062) (0.071) (0.100) (0.054) (0.047) (0.041) (0.025) WML − 0.012 − 0.011 − 0.070** 0.024 0.017 0.007 − 0.044* − 0.032 − 0.060** 0.083*** (0.031) (0.028) (0.035) (0.038) (0.034) (0.069) (0.026) (0.034) (0.024) (0.017) UMS 0.307*** 0.373*** 0.317*** 0.177 0.223*** 0.138 0.186** − 0.085* − 0.114*** − 0.236*** (0.080) (0.075) (0.096) (0.125) (0.058) (0.124) (0.085) (0.048) (0.043) (0.070) Constant 0.001 0.001 − 0.0003 0.001 0.0003 0.002* 0.0005 − 0.0003 0.00003 − 0.0003 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 83 83 83 83 83 83 83 83 83 83 Adjusted R20.964*** 0.961*** 0.955*** 0.957** 0.966*** 0.952 0.974*** 0.976 0.982** 0.981*** 66 J.Dobrick et al. Table 12 (continued) Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) ES Market 1.078*** 1.082*** 1.114*** 1.036*** 1.047*** 1.046*** 1.053*** 1.071*** 1.026*** 0.987*** (0.032) (0.018) (0.021) (0.018) (0.026) (0.022) (0.026) (0.018) (0.016) (0.019) SMB 0.345*** 0.503*** 0.302*** 0.399*** 0.275*** 0.291*** 0.213*** 0.030 − 0.119*** − 0.268*** (0.063) (0.103) (0.052) (0.053) (0.095) (0.082) (0.073) (0.043) (0.045) (0.045) HML − 0.049 − 0.061 0.035 − 0.087 − 0.102 − 0.004 0.042 0.163*** 0.098 − 0.053 (0.080) (0.065) (0.094) (0.066) (0.070) (0.082) (0.084) (0.049) (0.063) (0.042) WML − 0.142*** − 0.066 − 0.025 − 0.095* − 0.065 0.046 − 0.004 0.041 − 0.009 0.017 (0.049) (0.041) (0.053) (0.050) (0.051) (0.056) (0.034) (0.027) (0.022) (0.020) UMS 0.426*** 0.428*** 0.429*** 0.076 0.154 − 0.037 0.030 − 0.069 − 0.046 − 0.121*** (0.098) (0.128) (0.055) (0.071) (0.103) (0.054) (0.058) (0.066) (0.057) (0.044) Constant 0.002* 0.0003 0.0002 0.0004 0.002 0.001 0.0005 − 0.00004 − 0.001** 0.001 (0.001) (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 80 79 79 79 79 79 79 80 79 80 Adjusted R20.955*** 0.955*** 0.960*** 0.959 0.960* 0.964 0.967 0.974 0.975 0.987*** ESG Market 1.018*** 1.120*** 1.116*** 1.048*** 1.053*** 1.046*** 1.056*** 1.084*** 1.009*** 0.988*** (0.023) (0.038) (0.021) (0.021) (0.025) (0.019) (0.020) (0.039) (0.016) (0.015) SMB 0.281*** 0.489*** 0.379*** 0.415*** 0.198** 0.253*** 0.117 0.054 − 0.125*** − 0.238*** (0.060) (0.097) (0.078) (0.101) (0.080) (0.081) (0.086) (0.071) (0.046) (0.038) HML 0.008 − 0.044 − 0.023 − 0.058 0.012 − 0.107* 0.018 0.096** 0.115* − 0.029 (0.079) (0.068) (0.088) (0.061) (0.077) (0.058) (0.078) (0.042) (0.060) (0.034) WML − 0.128*** − 0.009 − 0.058 − 0.048 − 0.041 − 0.027 0.069*** − 0.009 0.014 0.011 (0.045) (0.042) (0.041) (0.059) (0.041) (0.026) (0.024) (0.025) (0.023) (0.014) UMS 0.453*** 0.386*** 0.353*** 0.148* 0.218** − 0.040 − 0.024 0.011 − 0.121** − 0.119** (0.156) (0.113) (0.098) (0.076) (0.095) (0.107) (0.091) (0.077) (0.051) (0.047) Constant 0.001* 0.001 − 0.001 0.002 0.0005 0.001* 0.001 0.0002 − 0.001 0.0001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 83 77 78 79 79 79 79 79 79 80 Adjusted R20.950*** 0.956*** 0.957*** 0.953 0.957** 0.967 0.968 0.974 0.978** 0.989*** This table shows the regression results of value-weighted decile portfolios for five different ESG dimensions. The utilized model is the Carhart four-factor model with an additional UMS factor accounting for each dimension of ESG *, ** and *** denote significance levels on the 10%, 5% and 1% significance levels, respectively. Significance level indications pertaining to the Carhart factors, UMS and constants are obtained from t-tests, and adjusted R2 indications are from F-tests for nested models. Newey–West standard errors are reported in parentheses 67ESG asrisk factor Table 13 Vigeo Eiris Europe: Carhart + UMS Decile (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Environmental Market 1.003*** 1.051*** 1.000*** 1.060*** 0.982*** 0.973*** 1.027*** 1.056*** 1.031*** 0.956*** (0.036) (0.038) (0.032) (0.020) (0.014) (0.023) (0.031) (0.043) (0.046) (0.029) SMB 0.243*** 0.148 0.231*** 0.033 − 0.016 − 0.097 − 0.054 − 0.125 − 0.078 − 0.288*** (0.057) (0.121) (0.079) (0.078) (0.081) (0.093) (0.125) (0.083) (0.080) (0.092) HML − 0.021 − 0.131 − 0.023 0.067 0.153*** 0.143* 0.068 − 0.093 0.135** − 0.013 (0.074) (0.091) (0.061) (0.076) (0.058) (0.083) (0.053) (0.071) (0.067) (0.049) WML 0.047 − 0.128 0.009 0.003 − 0.008 0.074 0.095*** − 0.078 0.110** − 0.114* (0.037) (0.094) (0.068) (0.044) (0.025) (0.055) (0.034) (0.049) (0.052) (0.058) UMS 0.116* 0.202*** 0.180*** 0.041 0.060 − 0.068 0.042 0.011 − 0.135*** − 0.074 (0.062) (0.064) (0.054) (0.056) (0.053) (0.050) (0.046) (0.053) (0.025) (0.079) Constant 0.001 0.001 − 0.001 0.001 − 0.001 0.001 − 0.0004 − 0.0003 − 0.001 0.002 (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 66 63 58 64 62 63 62 60 59 58 Adjusted R20.934** 0.928*** 0.940*** 0.950 0.961 0.952 0.951 0.966 0.959*** 0.942 Governance Market 1.021*** 1.042*** 0.994*** 1.010*** 1.016*** 0.986*** 1.009*** 1.044*** 1.029*** 0.983*** (0.021) (0.022) (0.019) (0.038) (0.030) (0.016) (0.034) (0.013) (0.022) (0.017) SMB 0.171*** 0.109 − 0.029 − 0.128 0.049 − 0.196*** − 0.108* − 0.147*** − 0.089 − 0.123** (0.050) (0.069) (0.051) (0.089) (0.097) (0.056) (0.059) (0.054) (0.061) (0.060) HML − 0.082 − 0.073* − 0.135** 0.081 0.195** 0.083 0.131*** 0.187*** 0.061 − 0.087** (0.053) (0.041) (0.053) (0.125) (0.078) (0.099) (0.049) (0.065) (0.054) (0.036) WML − 0.063 0.002 0.009 0.078* 0.112* − 0.035 0.048** − 0.046 0.014 − 0.013 (0.041) (0.028) (0.021) (0.041) (0.066) (0.047) (0.020) (0.047) (0.027) (0.025) UMS 0.305*** 0.183*** 0.368*** 0.118* 0.145* 0.206*** 0.043 − 0.377*** − 0.246*** − 0.374*** (0.052) (0.055) (0.067) (0.068) (0.088) (0.068) (0.057) (0.069) (0.055) (0.058) Constant 0.001 0.002* − 0.001 − 0.00000 − 0.0002 0.002* − 0.00004 − 0.0002 − 0.001 − 0.0003 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 64 62 62 62 62 62 61 62 59 60 Adjusted R20.957*** 0.966*** 0.966*** 0.950* 0.939** 0.958*** 0.957 0.952*** 0.968*** 0.956*** Social Market 1.022*** 1.044*** 1.025*** 1.019*** 0.966*** 0.992*** 0.976*** 1.062*** 1.003*** 1.005*** (0.021) (0.055) (0.036) (0.032) (0.031) (0.036) (0.042) (0.019) (0.017) (0.014) SMB 0.486*** 0.298*** 0.057 0.133 − 0.041 − 0.002 − 0.130* − 0.121** − 0.183*** − 0.256*** (0.070) (0.104) (0.060) (0.087) (0.101) (0.124) (0.079) (0.060) (0.046) (0.044) HML 0.112** − 0.178** − 0.065 − 0.116* − 0.069 0.026 0.035 0.056 0.0003 0.195*** (0.055) (0.071) (0.053) (0.060) (0.059) (0.074) (0.045) (0.048) (0.046) (0.041) WML − 0.061* 0.038 0.036 − 0.035 − 0.038 0.060* 0.095** − 0.005 0.031** − 0.028 (0.035) (0.046) (0.031) (0.029) (0.036) (0.033) (0.040) (0.027) (0.016) (0.022) UMS 0.167*** 0.281*** 0.232*** 0.053 0.026 0.065 − 0.021 − 0.018 − 0.067 − 0.168*** (0.041) (0.072) (0.052) (0.049) (0.053) (0.041) (0.060) (0.027) (0.050) (0.036) Constant 0.002*** − 0.002 0.002 0.001 0.001 − 0.002 − 0.001 0.001 0.001 − 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Observations 64 62 61 63 63 60 61 62 62 58 Adjusted R20.963*** 0.947*** 0.950*** 0.945 0.946 0.940 0.955 0.968 0.977* 0.981***