Abstract
This paper uses a text mining analysis to study the development of sustainable investing in the European and US insurance industry as reflected in their public reports from 2013 to 2018. The sample comprises 1215 annual, sustainability- and investment-related documents of 77 firms. We develop a dictionary with principles, criteria and terminologies as well as strategies, and differentiate between the quality of reports. Our results show that the number of firms referring to as well as the word count related to sustainable investing substantially increase over the sample period, and that insurers reporting about sustainable investing are on average significantly larger. We also find that European insurers report much more extensively on their sustainable investment practices as compared to US insurers in our sample. Most relevant in 2018 are references to general ESG criteria, followed by responsible investment and the Sustainable Development Goals. Top strategies mentioned were ESG integration and impact investing, whereby we observe that insurers evolve from mentioning one single towards multiple strategies over time. Finally, a regression analysis does not show a value-effect of sustainable investment-related keywords in reporting on Tobin’s Q, which may be due to the rather long-term investment perspective.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
In recent years, sustainable investment approaches have become increasingly important, as also reflected in more than USD 35 trillion sustainable investments globally at the beginning of 2020.Footnote 1 In this context, e.g. the UN-backed network “Principles for Responsible Investment” (PRI) with its about 3000 signatories with more than USD 100 trillion in assets under management in 2020 strongly promotes the incorporation of environmental, social and governance (ESG) criteria in the investment process.Footnote 2 The majority of the literature thereby suggests that investors can potentially benefit in terms of attractive long-term investment returns and enhanced risk management,Footnote 3 whereby most studies seem to make use of specific (and mostly external) data sources to identify sustainable investment activities.Footnote 4 Moreover, specifically for Europe, reporting requirements with respect to sustainability issues and sustainable investments are increasingly regulated, e.g. by the EU Directive 2014/95/EU on enhanced non-financial reporting as well as the EU Directive 2016/2341 for pension funds and life insurers with respect to accounting for ESG criteria in their business organization and reporting. One driving factor is also the ambitious sustainable finance-related Action Plan introduced by the European Commission (2018) to improve reporting, amongst other aspects.Footnote 5 In the US, in contrast, only the SEC’s (2010) guidance document on disclosing issues related to climate change seems to be one of the few comparable (and in this case rather topic-specific) regulations. As insurance companies are among the largest institutional investors with an asset volume of about EUR 10 trillion in Europe alone,Footnote 6 and given that Europe and the US are the two largest sustainable investment markets,Footnote 7 their approach to sustainable investing should be of great interest for various stakeholders. However, the investment policy of institutional investors is still largely unexplored, as pointed out by Cunha et al. (2021).
Against this background, the aim of this paper is to contribute to the literature by providing a detailed study of sustainable investing in the European and US insurance industry, as reflected in insurers’ public reports from 2013 to 2018. To conduct such an in-depth analysis, we first develop a sustainable investing keyword dictionary and then use a text mining approach applied to annual, sustainability- and investment-related reports and documents. Our sample comprises 77 large-cap European and US insurance companies with a market capitalization share of about 56% of the initial sample, retrieved from Refinitiv Eikon and Datastream. This results in 1215 annual, sustainability- and investment-related reports and documents based on which we aim to gain insight regarding sustainable investing of European and US insurers. The application of the text mining approach avoids reliance on external data providers, while capturing minor and major prevalent principles, criteria, terminologies and strategies in this context for a large number of firms. We thus contribute to the ongoing discussion in the academic and practitioner-oriented literature on sustainable investing and provide a first in-depth reporting-based analysis of the development and status in the insurance industry as a major (sustainable) institutional investor.
The remainder of this paper is structured as follows. Section “Data sample” presents the data sample; Sect. “Methodology” describes the text mining approach with the developed sustainable investment dictionary; Sect. “Results” presents the results; and Sect. “Summary” concludes.
Data sample
We use the Refinitiv Eikon database and start with all firms from the Thomson Reuters Business Classification (TRBC) sector ‘Insurance’ located in the European Union (including the UK) and the US with an available market capitalization at the end of 2018 in Datastream. As publicly traded companies are more likely to report their sustainable investment approach, we consider large cap firms with more than USD 1 billion market capitalization in 2018 and emitted ordinary shares. Finally, we exclude nine firms with a missing market capitalization in Datastream between 2013 and 2018 and one insurer with a missing annual report. We also check their business description in their annual reports, amongst others, and exclude insurance brokers, for instance.Footnote 8 This approach leads to 48 US and 29 European insurance companies with a total market capitalization of USD 884 billion, representing 56.4% of the market capitalization of the initial sample.
We use a text mining approach to identify whether insurance companies report about sustainable investing starting from 2013 to 2018, which is based on (group) annual reports, reviews and summaries (Annual Reporting “AR”) and sustainability reports (Sustainability Reporting “SR”). Besides the AR and SR categories, further reports, presentations, documents and policies related to corporate social responsibility (CSR), ethics or to the sustainable investment approach introduced by insurers represent a third category in this analysis (Sustainability-, Investment-related and Others “SIO”). As these (unaudited) documents may be less reliable concerning the quality of information compared to AR and SR, we consider the SIO category in additional analyses in the Appendix. The majority of these files is available on corporate websites in PDF format and are related to the six-year sample period 2013–2018.Footnote 9 ‘Annual’ and ‘ESG Disclosure’ filings are reviewed in the Refinitiv Eikon database for each company as well. The SEC’s EDGAR platform is also used to search for missing forms and amendments.Footnote 10
Furthermore, in line with studies examining CSR disclosures (see, e.g., Dhaliwal et al. 2011), we screen the following external sources for supplementary reports and documents: the Global Reporting Initiative (GRI) Sustainability Disclosure Database, CorporateRegister.com, the UN Global Compact participants archive (‘Communication on Progress’) as well as the disclosures of signatory companies of the Principles for Sustainable Insurance (PSI). An internet search using the first three Google pages is additionally performed for SR and SIO documents of the considered firms, e.g. including the keywords “responsible”, “investment”, “policy” or document names, if it is not clear whether a firm published it for a given year.Footnote 11 We do not consider stand-alone reports and documents published by asset managers (possible disclaimer issues), related to the Climate Risk Disclosure Survey (regionally limited), the Carbon Disclosure Project, PRI Transparency reports as well as summary sheets with key performance indicators (KPI) or Yes/No questions (inconvenient (query-response) structure). Strategic documents with focus beyond the sample period (forward-looking) and snapshots from corporate websites (problematic traceability and classification), created by and available in the Refinitiv Eikon ESG database (e.g. ‘ESG Web-based Reporting’), are also not included in the analysis. While either an annual report or a 10-K form must be available for each company and year, sustainability reporting is not a prerequisite in order to remain in the sample.Footnote 12
This approach leads to a total number of 619 documents in the AR category (e.g. form-10K/A, (financial or annual) reviews, management reports, annual letters, especially relevant for US insurers) and 250 files in the SR category. The SIO category comprises 346 documents used for additional analyses in the Appendix. Hence, 1215 files are considered in total, including 131 copies of documents or policies in line with the aforementioned approach. Table 6 in the Appendix summarizes the data collection approach and the number of reports and documents per category and region.
Methodology: a text mining approach
We use the data mining software RapidMiner, which allows us to evaluate a large number of reports with a reduced error susceptibility and a standardized procedure by considering root words,Footnote 13 while introducing other constraints such as paraphrasing or syllabification (see Heidinger and Gatzert 2018). Corrupted documents as well as a possible non-consideration of relevant context or punctuation represent further limitations. Besides examining the development over time in terms of absolute numbers, the text mining tool also enables us to take into account the development in relation to the total amount of words per report and in relation to all references to investments as is similarly done in Heidinger and Gatzert (2018) with an application to reputation risk management.
A multitude of terminologies and approaches has emerged in the literature capturing different perceptions of investing in a sustainable manner (see, e.g., Sandberg et al. 2009).Footnote 14 In line with GSIA (2019, p. 3) as the global network of sustainable investment organizations, we define sustainable investing as a generic concept encompassing all other linked terms, strategies and concepts such as socially responsible investing and responsible investing and which includes ESG criteria in the investment management and selection (see also GSIA 2021). As we study European and US insurance companies, we mainly revert to the relevant and most recent publications of the respective investment forumsFootnote 15 on the status of sustainable investing, which also offer definitions on sustainable investment strategies (see Eurosif 2018; USSIF 2018, 2020; GSIA 2019, 2021). Moreover, we also take into account the definitions of the PRI Association (2018) as well as academic publications from the research field. With regard to the latter, this also includes literature reviews and publications on emerging themes in the sustainable investment context (see, e.g., Inderst and Stewart 2018; Talan and Sharma 2019; Daugaard 2020).
Based on this, Table 1 summarizes our sustainable investment dictionary with important principles, criteria and terminologies as well as strategies from the sustainable investment theory and practice by offering the respective keyword(s), a contextual definition and the related source(s). As can be seen from Table 1, some terms have a different but still somewhat close interpretation according to the respective sources (e.g. sustainable investment, sustainable and responsible investment, responsible investment), which is why we later also provide aggregate statistics. While our dictionary generally matches with the previously mentioned publications, it still represents a selection of keywords and will thus not be fully exhaustive. Moreover, we do not consider the large variety of specific ESG- or sustainability-related themes and issues (e.g. “renewable energy” or “diversity”) or generic words that are frequently used in reporting or in a different context (e.g. “long-term investing” or “divesting”). We also exclude words and concepts with different meanings, such as “community investment” in the impact investment context (see, e.g., Tsang et al. 2009; Rowe et al. 2014; GSIA 2019).
Before we aggregate the counts of the keywords for each strategy and terminology as well as for the principles and criteria from Table 1, we perform a subtraction of double counts.Footnote 16 We also identify three common wordings (“impact investment income”, “impact investment returns” and “sustain investment income”), in particular in the context of US firms, which would lead to false positive results for impact investing and sustainable investment, respectively, and are thus subtracted. Within the scope of the engagement strategy, active ownership represents another issue that has to be considered in more detail, as it is used as a Sector specific Aspect for the financial services sector in the Social Category (Sub Category: Product Responsibility) within the GRI (2013, p. 8) framework.Footnote 17 Insurers which follow the GRI standards might thus list this Sector specific Aspect in their GRI content index without providing information on the application, for instance. We thus review all documents with a positive hit for active ownership and again adjust the word counts and number of identified firms if the insurer negates the application or if no further information is provided.
Text mining results: sustainable investing in the US and European insurance industry
Development of references to sustainable investing in annual and sustainability reports in US and European insurance companies from 2013 to 2018
As a first indication for the development of sustainable investing in the US and European insurance industry, Fig. 1 displays the proportion of examined keywords from our sustainable investment dictionary in Table 1 in relation to the total number of words (tokens) in the reports (Fig. 1a) and in relation to the total number of investment references (Fig. 1b) over the sample period. We further distinguish between the region and the type of report, thereby focusing on the annual (recurring) report types AR (left y-axis) and SR (right y-axis).
Figure 1 indicates that sustainable investing has become increasingly important in relative terms and that there are strong differences depending on the type of report and region. For instance, it seems that US annual reports are hardly used to inform about sustainable investing, which remains rather stable over time in both Fig. 1a and b (lowest solid line with squares). However, US insurers increasingly use their sustainability reports to present information about their sustainable investment approach. In Europe, in contrast, annual reports do appear to serve as an instrument to report about a sustainable investment approach in addition to sustainability reports, which has been strongly increasing over time. These observations remain consistent in Fig. 1b.
Table 2 presents more detailed descriptive statistics for the years 2013 and 2018. In line with the previous observations, we do not only see an increase in the number of keyword hits for almost all categories in Table 2a and b but also in the number of firms using these keywords.
Looking at Table 2a, most relevant in 2018 are references to the Sustainable Development Goals (400 word counts), closely followed by responsible investment (374 word counts) and ESG criteria (250 word counts; 514 word counts for all ESG references combinedFootnote 18). This is in line with the references to the Principles for Responsible Investment (107 word counts),Footnote 19 an initiative that strongly builds on the term responsible investment and ESG criteria. Moreover, the PRI network also promotes and supports (the achievement of) the SDGs as sustainability issues among its stakeholders.Footnote 20 However, when looking at the number of firms referring to the principles, criteria and terminologies, we find a slightly different ranking, with the highest number of firms (44 of 77) in the sample mentioning ESG criteria, while the second highest number of firms (38) refers to responsible investment. The high word count for the SDGs (by 28 firms) appears to be driven by firms which comprehensively report about them, as reflected in the largest standard deviation and maximum.
Whereas the SDGs present strong growth rates both in terms of the word counts and number of firms since their official introduction in 2015,Footnote 21 keyword hits related to the PRI and PSI show an overall smaller or no increase. However, when looking at the number of firms, the text mining results still indicate a strong increase in awareness for the PRI. We find that 23 insurers in our sample mention the PRI in 2018, starting from eleven insurers in 2013. The number of firms mentioning the PSI is smaller with initially seven insurers in 2013 and ten in 2018. Overall, these results indicate an increasing relevance of initiatives such as the PRI or PSI among insurers as well as a growing relevance of sustainability risks and opportunities in the insurance industry in general (see also Gatzert et al. 2020; Gatzert and Reichel 2022).
With regard to further terminologies in Table 2a, ESG investing and ESG incorporation show a large increase in the number of references by firms, but the number of word counts is relatively small, which might be explained by further references to other ESG-related terms and strategies in the reports and documents. While the number of references to sustainable investment more than doubles over time, references to sustainable and responsible investment and sustainable, responsible and impact investing (included in the category other) as promoted by Eurosif (2018) and USSIF (2018), respectively, are hardly found or not at all. The same can be concluded for the Equator Principles (included in other).
Concerning the keyword counts and number of firms referring to sustainable investment strategies in Table 2b, most often referenced in 2018 are ESG integration (19 firms with 155 hits) and impact investing (19 firms with 84 hits), followed by references to green bonds (16 firms with 52 hits). While insurance companies report less about screening and engagement, we find hardly any references to sustainability themed investing as well as to social bonds and sustainability bonds. With regard to the latter strategies, it appears that these thematic-driven investments are rather at the beginning of their development in the insurance industry, being mentioned either in 2017 or 2018 for the first time.
As the PRI play an increasing role among institutional investors while also promoting the application of sustainable investment strategies, we additionally calculate Pearson’s and Spearman’s correlation coefficients and find statistically significant and positive correlations between firms referring to the PRI and their amount of word counts for each strategy. Green bonds (Pearson’s) and ESG integration (Spearman’s) show the highest correlation coefficients in this context.
Differences in reporting in US and European insurance companies with respect to sustainable investing and the impact of regulation
We next study differences between European and US insurers regarding the number of references to sustainable investment-related keywords in Table 3 and regarding the number of firms with such references in Table 4. Starting with the principles, criteria and terminologies in Table 3a, the results again indicate strong regional discrepancies within the sample. In 2018, European insurers with positive hits in regard to keywords in Table 1a have a total keyword count of 1193, which is close to four times the keyword count of US insurers (307 in total). In 2013, US insurers have only 24 out of 463 positive hits. Furthermore, the respective rankings differ. When looking at the number of firms in Table 4a, it appears that in our sample, ESG criteria and the PRI did not play a large role in the US insurance industry until 2016. It also seems as if the PSI are hardly disseminated in the US, with the first two references in 2017 and one consecutive hit in 2018 attributed to only one firm. With respect to investment strategies in Tables 3b or 4b, besides ESG integration, the impact investing strategy is mentioned by seven US firms in 2018, while the strategies engagement and screening do not play a relevant role in the US sample.
One potential driver for the general increase in the reporting on sustainable investing, especially for European insurers, is likely regulation as described in the introduction, e.g. the EU Directive 2014/95/EU from 2014, effective 2017. This so-called Non-Financial Reporting Directive (NFRD) focuses on more corporate transparency in regard to non-financial and diversity reporting. In the US, in contrast, it seems as if more transparency in regard to the consideration of sustainable investment has not yet been in the focus of regulation. On a global scale, the adoption of the Paris Agreement by 196 parties in December 2015 might also represent a key event in this context, as the mobilization of the financial market participants constitutes a central element as an enabler to reduce emissions and to become more resilient to climate change.Footnote 22
To study the impact of these regulations in more detail, we use a paired t-test in order to test whether there is a significantly higher number of references to keywords in our sustainable investment dictionary, i.e. a higher word count in Table 3, for the year 2017 as compared to the year 2016 for European insurers, and for the year 2016 compared to the year 2015 for the whole sample. The results show that on average, the word count is significantly higher in the year 2016 after the adoption of the Paris Agreement for the whole sample (x̅Total15 = 9.857 and x̅Total16 = 14.195; x̅Total16 − x̅Total15 = 4.338; p value = 0.001). Moreover, on average, the word count is also significantly higher in the year 2017 as the first effective year of the NFRD for the European sample (x̅EU16 = 33.241 and x̅EU17 = 44.759; x̅EU17 − x̅EU16 = 11.518; p value = 0.003).Footnote 23
Using a two-sample t-test and considering all firm-year observations, we further find statistically significant differences in means between European and US insurers at the 1% level for ESG integration, impact investing, green bonds, screening and engagement, with higher means for EU insurers. The same can be concluded with respect to the number of firms referring to these strategies.Footnote 24
We also note that a relevant number of European insurers in our sample still does not report at all about these concepts. Figure 2a illustrates this issue in more detail by presenting the number of firms per region with at least one positive hit from our sustainable investment dictionary in Table 1. Out of all 29 European insurers in the sample, in 2013, eight insurers do not mention sustainable investment-related keywords at all, and in 2018, there is still one UK insurer without any keyword hit. In the US, in 2018, we have 28 out of 48 insurers without any positive hit from the sustainable investment dictionary, compared to 42 “non-reporting” US insurers in 2013.
To get further insight into the potential depth of the application of a sustainable investment approach, Fig. 2b displays the number of US and EU insurers that have at least one keyword hit in Table 1a and at least one keyword hit in Table 1b. The results show an increase from 13 firms (out of 77) in 2013 to 34 in 2018, with considerably less US than European insurers. All these insurers use and combine keywords from both Table 1a and b, e.g. by offering a sustainable investment section which includes the applied strategies such as ESG integration.
The impact of firm size
Apart from regional differences, we also study whether firms with and without at least one keyword hit from the sustainable investment dictionary (Fig. 2a) differ in terms of their firm size, and analogously for firms in Fig. 2b.
Table 5 shows group differences in means and medians as well as Pearson’s and Spearman’s correlation coefficients. The results show that insurers reporting about sustainable investing as reflected in Fig. 2a and b are significantly larger on average than those that do not report accordingly, and that there is a statistically significant and positive relation (in terms of correlations) between insurers that do report about sustainable investing and their firm size. The differences in means and medians are even higher for firms identified in Fig. 2b.
The number of sustainable investment strategies applied in US and European insurance companies according to their annual and sustainability reports
We further investigate the firms from Fig. 2a (i.e., at least one keyword hit from Table 1) in regard to the number of reported sustainable investment strategies (listed in Table 1b, i.e. at most six) depending on the region and year in Fig. 3. We can observe that the number of firms applying multiple sustainable investment strategies increases over time. While the majority of insurers (28 out of 34) report about one to three different strategies in 2018, five firms mention four strategies, and one insurer even refers to all six strategies. In contrast, 13 of the 14 identified insurers in 2013 mention one to three different strategies, while the maximum number of strategies applied is four by one insurer. We also note that the majority of firms cite only one strategy in their first year and that the results are driven by European insurers in the sample. Overall, the analysis of the reports suggests that insurers develop their sustainable investment approaches further by applying multiple strategies, which is also recommended by the PRI Association (2018) (e.g. to combine screening and engagement) and also observed by the GSIA (2021) for the overall investment community.
Differences between subsectors
Differences depending on the subsectors life & health (LH), multiline insurance (ML) and property & casualty (PC) are summarized in Table 7 in the Appendix.Footnote 25 When looking at the number of firms per strategy (see Table 7a), we find that the results are driven by life & health insurers, followed by multiline and property & casualty insurers. With respect to word counts (see Table 7b), multiline insurers show the strongest relative increase over time with a factor of around eight. Although around half of our sample firms are property & casualty insurers, we only find a small number of mainly European firms from this subsector referring to sustainable investment strategies in the first years. The number of reported strategies then particularly increases in the years 2017 and 2018. One reason for this result might be the fact that most US companies are property & casualty insurers (more than 60% (37%) within the US sample (overall sample)), and that according to previous observations, US insurers in the sample mention considerably less sustainable investing activities in their reports and documents (e.g. between 2014 and 2017, no US property & casualty insurer reports about sustainable investment strategies). In addition, aforementioned regulation initiatives and the long-term investment perspective of life & health insurers might be reasons for the stronger focus on sustainable investing, specifically with focus on ESG integration, impact investing, including bonds, and engagement.
Robustness checks: including sustainability-, investment-related and other documents
Finally, when adding the documents from the “Sustainability-, Investment-related and Others” (SIO) category to previous analyses as a robustness check, we find that the majority of the already identified firms uses these types of documents to further inform their stakeholders about their sustainable investment approach. Many word counts substantially increase, e.g. in 2018, the word counts for the PSI as well as for ESG integration are even higher within the SIO category than in the AR and SR categories combined (see Table 8 in the Appendix). The additional growth in the number of identified firms reporting about sustainable investing is again mainly driven by European insurers (see Table 9 in the Appendix). Moreover, Fig. 4a in the Appendix shows the additional number of European and US insurers in the sample with at least one keyword hit from the sustainable investment dictionary in Table 1 based on the supplementary consideration of the SIO category. We find that only two European firms are newly identified compared to Fig. 2a (the increase in 2015 and 2016 is based on the same firm). The remaining firms already have references to the principles, criteria and terminologies in their AR and/or SR and further disseminate information on their investment strategies by using the SIO documents. This also explains the increase in Fig. 4b in the Appendix compared to Fig. 2b.
The value-relevance of sustainable investment-related keywords
We conclude with a panel regression model with fixed effects to study the value-effect of sustainable investment-related keywords following e.g. Hoyt and Liebenberg (2011), Heidinger and Gatzert (2018) and Gatzert and Reichel (2022) by using Tobin’s Q as the dependent variable. We thereby measure the variable of interest as the natural logarithm of all keyword counts from our sustainable investment dictionary (see Tables 1 and 3), which are identified in annual and sustainability reports per firm. Following previous literature, we further control for size, leverage and return on assets. Overall, the results do not show a statistically significant relation between sustainable investment-related keywords and the dependent variable Tobin’s Q, which might be explained by the foundation of sustainable investing as a more long-term investment approach, where the implementation can be costly, while benefits might materialize over a longer time horizon.
Summary
The aim of this paper is to gain insight regarding sustainable investing activities in the European and US insurance industry. This topic is of high relevance especially for insurers as one of the largest institutional investors that also face an increasing regulatory pressure in this regard. We therefore develop a sustainable investment dictionary with keywords that take into account sustainable investment strategies as well as principles, criteria and terminologies. Based on this dictionary, we apply a text mining process to 1215 annual, sustainability- and investment-related reports and documents of 77 large-cap European and US insurance companies from 2013 to 2018.
First, we find that references to our sustainable investment dictionary have been strongly increasing in annual and sustainability reports over the sample period, and that insurers reporting about sustainable investing are significantly larger on average than those without keyword hits. Second, (the more regulated) European firms in the considered sample exhibit a much more extensive reporting about sustainable investment concepts and principles than US insurers. Third, while the rankings show similar tendencies across the two regions, we also observe differences. In regard to the number of firms and the related word count in the last sample year 2018, most often cited concepts are ESG criteria together with responsible investment and the Sustainable Development Goals, whereby the SDGs exhibit the strongest growth rates since their official introduction in 2015. Furthermore, while the Principles for Responsible Investment are also highly cited, the Principles for Sustainable Insurance are ranked lower with hardly any reference from the US. Fourth, with respect to investment strategies, ESG integration and impact investing are most often mentioned, while engagement and screening did not play a substantial role specifically in the US. We also find that more recently, an increasing number of insurers refers to multiple strategies and that especially life & health insurers refer to strategy-related keywords. Fifth, when including documents from our third reporting category “Sustainability-, Investment-related and Others”, the robustness results show that almost all firms are already identified based on their annual and sustainability reporting. Finally, a panel regression with fixed effects does not show a value-effect of sustainable investment-related keywords used in insurers’ annual or sustainability reports on Tobin’s Q.
One limitation of this approach is that firms may not fully disclose their sustainable investment strategies or only refer to their (anonymized) external asset manager. Moreover, the interpretation is restricted in that the context cannot be taken into account in an automated text mining process. However, the text mining approach allows us to analyze a large number of reports (in our case 1215 files), which would not be possible manually. Using this large database, it allows in-depth insight into the relevance of specific strategies, principles and terminologies that insurers cite in their reports and published documents, and their development over time.
Based on our analysis, several avenues for future research can be derived. To challenge the reported actions, to identify undisclosed strategies or to further evaluate implemented practices, a survey could be performed as an alternative promising approach as discussed, for instance, by Hoyt and Liebenberg (2011) or Krueger et al. (2020). Future research could also make use of surveys to identify insurance companies with major asset management operations or ESG units and then examine the relationship with sustainable investment approaches in more detail. Besides conducting surveys, interviews or case studies can be conducted to study the relevance of national or state-related ESG regulations in Europe or the US, which could be further specified for subsectors. Finally, following more recent publications such as Talan and Sharma (2019) and Daugaard (2020), our analysis could be extended to emerging markets, which have not been in the focus of research yet, and a more detailed analysis of the value-effect of sustainable investment-related reporting would be of interest as well.
Overall, against the background of the substantial investment amounts in the insurance industry and benefits for the environment and society as well as potentially attractive long-term returns as suggested by the literature, sustainable investment strategies should be a vital consideration of insurers. In addition, as transparency plays a key role in assessing risks inherent in an insurer’s investment portfolio and since pressure increases by various stakeholders of insurance companies (regulation, NGOs, customers), we expect that the absolute and relative proportion of disclosing information on sustainable investing will increase over time. Our observations suggest that insurers increasingly develop more elaborated investment strategies. Instead of simply excluding specific sectors for instance, insurance companies will likely apply multiple strategies in the future.
Notes
This represents more than one third of all assets being managed by investment professionals from the regions included by the Global Sustainable Investment Alliance (GSIA 2021).
See https://www.unpri.org/download?ac=10948, accessed 09 Jul 2021.
The impact of taking into account sustainability or ESG factors in asset management on financial performance has been extensively researched as well (see, e.g., Revelli and Viviani 2015, for a review). Besides focusing on firm- and investment- or portfolio-levels, an increasing number of articles also specifically focuses on institutional investors (see, e.g., Bengtsson 2008; Scholtens and Sievänen 2013; Dyck et al. 2019).
The European Commission’s (2018, p. 2) “Action Plan: Financing Sustainable Growth” aims to “1. reorient capital flows towards sustainable investment in order to achieve sustainable and inclusive growth; 2. manage financial risks stemming from climate change, resource depletion, environmental degradation and social issues; and 3. foster transparency and long-termism in financial and economic activity”. With regard to the second point, the Solvency II review and its potential contribution is explicitly mentioned in this context, for instance.
See https://www.insuranceeurope.eu/statistics, accessed 09 Jul 2021.
According to the GSIA (2021), Europe and the US account for more than 80% of all sustainably invested assets on a global scale from 2018 to 2020.
We exclude four insurance brokers (Aon PLC, Willis Towers Watson PLC, Arthur J Gallagher & Co and Brown & Brown Inc), one professional services firm including insurance brokerage (Marsh & McLennan Companies Inc), two conglomerates (Berkshire Hathaway Inc and Loews Corp) and one firm with a missing annual report in English (Wuestenrot & Wuerttembergische AG) from the final sample as well. Note that the exclusion of Berkshire Hathaway Inc has a large impact on the remaining market share of our sample with a market capitalization of USD 502 billion in 2018, which represents around one third of the initial sample.
Sustainability reports are considered as an additional information source possibly extending annual reports and 10-K forms (see Dhaliwal et al. 2011). Typical labels for this kind of report are, for instance, Corporate (Social) Responsibility Report, Integrated Report, GRI Table or Content Index, Corporate Citizenship Report, ESG Report or Sustainability Report. The inclusion of a (reporting package of) document(s) in the SR category is based on the name of the report as well as on information from the corporate websites and external sources. Policies or certain statements from the SIO category, which had been published before 2013 or in one of the sample years, are assumed either to be in place in the following year(s) until the last sample year or until a new or updated version is found for the specific insurance company. A missing or non-recurring consideration would affect the reproduction of the actual investment strategies of the firms.
A large number of US firms publishes comprehensive annual reports, which typically consist of an annual letter to the shareholders, financial highlights and (content of) the 10-K form. Alternatively, a short annual report, review or a stand-alone letter is prepared without the 10-K form. We primarily include comprehensive annual reports or at least 10-K forms and add further (qualitative) reports for US insurers, if applicable.
As a clear reporting period or publication date is not always available or straightforward, we look at the document for information regarding the period or date as well as at the databases. If we do not find a clear statement, we search for further indicators (e.g. document name, copyright date, (financial) data, references from or to other documents). The year allocation is also based on the company- or document-specific publication cycle. In most of the cases, we allocate the documents to one single year, based on the aforementioned approach. However, while annual reports are always published for and assigned to one single year, other reports and documents may focus on multiple years, e.g. considering the years 2015 and 2016. In specific cases, we thus assign the document to two years, if, for instance, both years are (partially) covered by the document and a publication cycle is not available or being the last report within a cycle (e.g. four out of 250 documents from the SR category are used in more than one year in line with this approach).
In contrast to annual reports with due dates, firms (sometimes) tend to publish sustainability-related reports and documents with (substantial) delay. Moreover, as our sample shows, some firms even combine multiple years in one sustainability report. We thus use 2018 as our cut-off year in order to provide a high level of consistency with the aim to reduce data availability bias, i.e. missing reports and documents for firms in the sample (specifically in the SR and SIO category), together with our comprehensive data collection approach.
‘Stemming’ as process to generate root words is applied on the following words: ‘invest’, ‘sustain’, ‘integ’, ‘excl’, ‘engag’, ‘screen’, ‘bond’, ‘norm’, ‘approach’, ‘incorp’, ‘principle’, ‘responsibl’, ‘value’, ‘ethical’, ‘goal’, ‘income’ and ‘return’. These words are examined in combination with other words, such as ‘green’ ‘bond’ or ‘sustain’ ‘invest’.
For instance, while the PRI-promoted concept and term responsible investment does not require a moral return and is defined “as a strategy and practice to incorporate […] [ESG] factors in investment decisions and active ownership” (https://www.unpri.org/download?ac=10223, accessed 21 Jun 2021), practitioners and academics claim that socially responsible investments, for instance, consider moral aspects as a possible fourth dimension (see, e.g., Hebb et al. 2014). Following the PRI Association (2018), the ESG criteria refer to issues on the status and resilience of environment and nature (E), on social and civic rights, concerns and prosperity (S) as well as on corporate governance on investment level (G).
For instance, the term ‘principle* for responsibl* invest*’ encloses ‘responsibl* invest*’. As the latter also includes the counts for the PRI term, the ‘responsibl* invest*’ count is adjusted, respectively.
This includes “[v]oting policy(ies) applied to environmental or social issues” (GRI 2013, p. 36) and two related indicators.
This includes word counts for environmental, social and governance (250), ESG investing (48), ESG incorporation (42), ESG integration (155) and ESG screening (19; included in screening with 39 word counts) in 2018.
Note that there is no double counting, as the number of relevant word counts for Principles for Responsible Investment is subtracted from the counts for responsible investment, for instance.
See https://www.unpri.org/sustainability-issues/sustainable-development-goals, accessed 05 Jul 2021.
See https://www.un.org/sustainabledevelopment/development-agenda/, accessed 05 Jul 2021. First hits in 2013 and 2014 refer to an outlook on the introduction of the SDGs by two different insurers.
See https://unfccc.int/process-and-meetings/the-paris-agreement/the-paris-agreement, accessed 01 Mar 2022; https://unfccc.int/sites/default/files/english_paris_agreement.pdf, accessed 01 Mar 2022.
We also apply a Wilcoxon signed-rank test, respectively, which generally confirms our results (European sample: p value = 0.006; whole sample: p value = 0.001).
The number of observations for European insurers is nEU = 174 and the number of observations for US insurers is nUS = 288. For instance, the difference in means for the word count of impact investing is 0.716 (x̅IIWEU = 1.167 and x̅IIWUS = 0.451; p value = 0.003) and 0.174 for the number of firms referring to impact investing (x̅IIFEU = 0.264 and x̅IIFUS = 0.090; p value = 0.000). We also apply a Wilcoxon rank-sum test as well as a chi-square test of independence for binary variables, which generally confirm our results.
Our sample consists of 20 life & health insurers (13 US, 7 European), 18 multiline insurers (6 US, 12 European) and 39 property & casualty insurers (29 US, 10 European).
References
Bengtsson, E. 2008. Socially responsible investing in Scandinavia - A comparative analysis. Sustainable Development 16 (3): 155–168.
Brooks, C., and I. Oikonomou. 2018. The effects of environmental, social and governance disclosures and performance on firm value: A review of the literature in accounting and finance. The British Accounting Review 50 (1): 1–15.
Cunha, F.A.F.d.S., E. Meira, and R.J. Orsato. 2021. Sustainable finance and investment: Review and research agenda. Business Strategy and the Environment 30 (8): 3821–3838.
Daugaard, D. 2020. Emerging new themes in environmental, social and governance investing: A systematic literature review. Accounting & Finance 60 (2): 1501–1530.
Dhaliwal, D.S., O.Z. Li, A. Tsang, and Y.G. Yang. 2011. Voluntary nonfinancial disclosure and the cost of equity capital: The initiation of corporate social responsibility reporting. The Accounting Review 86 (1): 59–100.
Dyck, A., K.V. Lins, L. Roth, and H.F. Wagner. 2019. Do institutional investors drive corporate social responsibility? International evidence. Journal of Financial Economics 131 (3): 693–714.
European Commission. 2018. Action plan: Financing sustainable growth. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018DC0097&from=EN. Accessed 10 Jul 2021.
European Sustainable Investment Forum (Eurosif). 2018. European SRI study 2018. https://www.eurosif.org/wp-content/uploads/2021/10/European-SRI-2018-Study.pdf. Accessed 05 Mar 2019.
Gatzert, N., and P. Reichel. 2022. Awareness of climate risks and opportunities: Empirical evidence on determinants and value from the U.S. and European insurance industry. The Geneva Papers on Risk and Insurance - Issues and Practice 47 (1): 5–26.
Gatzert, N., P. Reichel, and A. Zitzmann. 2020. Sustainability risks & opportunities in the insurance industry. Zeitschrift für die gesamte Versicherungswissenschaft 109 (5): 311–331.
Giese, G., L.E. Lee, D. Melas, Z. Nagy, and L. Nishikawa. 2019. Foundations of ESG investing: How ESG affects equity valuation, risk, and performance. The Journal of Portfolio Management 45 (5): 69–83.
Global Reporting Initiative (GRI). 2013. G4 sector disclosures: Financial services. https://www.globalreporting.org/search/?query=Financial+services. Accessed 03 Sept 2021.
Global Sustainable Investment Alliance (GSIA). 2019. Global sustainable investment review 2018. http://www.gsi-alliance.org/wp-content/uploads/2019/03/GSIR_Review2018.3.28.pdf. Accessed 19 Jan 2020.
Global Sustainable Investment Alliance (GSIA). 2021. Global sustainable investment review 2020. http://www.gsi-alliance.org/wp-content/uploads/2021/08/GSIR-20201.pdf. Accessed 18 Oct 2021.
Hebb, T., C. Louche, and H. Hachigian. 2014. Exploring the societal impacts of SRI. In Socially responsible investment in the 21st century: Does it make a difference for society?, ed. C. Louche and T. Hebb, 3–20. Bingley: Emerald.
Heidinger, D., and N. Gatzert. 2018. Awareness, determinants and value of reputation risk management: Empirical evidence from the banking and insurance industry. Journal of Banking & Finance 91: 106–118.
High-Level Expert Group on Sustainable Finance (HLEG). 2018. Financing a sustainable European economy: Final report 2018. https://ec.europa.eu/info/sites/info/files/180131-sustainable-finance-final-report_en.pdf. Accessed 15 Jan 2019.
Hoyt, R.E., and A.P. Liebenberg. 2011. The value of enterprise risk management. Journal of Risk and Insurance 78 (4): 795–822.
Inderst, G., C. Kaminker and F. Stewart. 2012. Defining and measuring green investments: Implications for institutional investors' asset allocations. OECD Working Papers on Finance, Insurance and Private Pensions, No. 24, OECD Publishing.
Inderst, G., and F. Stewart. 2018. Incorporating environmental, social and governance (ESG) factors into fixed income investment. World Bank Group Publication, Washington, DC.
International Capital Market Association (ICMA). 2021a. Green Bond Principles: Voluntary process guidelines for issuing green bonds (June 2021). https://www.icmagroup.org/assets/documents/Sustainable-finance/2021-updates/Green-Bond-Principles-June-2021-140621.pdf. Accessed 20 Jun 2021.
International Capital Market Association (ICMA). 2021b. Social Bond Principles: Voluntary process guidelines for issuing social bonds (June 2021). https://www.icmagroup.org/assets/documents/Sustainable-finance/2021-updates/Social-Bond-Principles-June-2021-140621.pdf. Accessed 01 Jul 2021.
International Capital Market Association (ICMA). 2021c. Sustainability Bond Guidelines (June 2021). https://www.icmagroup.org/assets/documents/Sustainable-finance/2021-updates/Sustainability-Bond-Guidelines-June-2021-140621.pdf. Accessed 01 Jul 2021.
Krueger, P., Z. Sautner, and L.T. Starks. 2020. The importance of climate risks for institutional investors. The Review of Financial Studies 33 (3): 1067–1111.
Majoch, A.A.A., A.G.F. Hoepner, and T. Hebb. 2017. Sources of stakeholder salience in the responsible investment movement: Why do investors sign the Principles for Responsible Investment? Journal of Business Ethics 140 (4): 723–741.
Malik, M. 2015. Value-enhancing capabilities of CSR: A brief review of contemporary literature. Journal of Business Ethics 127 (2): 419–438.
Margolis, J.D., and J.P. Walsh. 2003. Misery loves companies: Rethinking social initiatives by business. Administrative Science Quarterly 48 (2): 268–305.
PRI Association. 2018. PRI reporting framework: Main definitions (November 2018). https://d8g8t13e9vf2o.cloudfront.net/Uploads/d/t/z/maindefinitionstoprireportingframework_127272.pdf. Accessed 17 Jun 2021.
Revelli, C., and J.-L. Viviani. 2015. Financial performance of socially responsible investing (SRI): What have we learned? A meta-analysis. Business Ethics: A European Review 24 (2): 158–185.
Rowe, A.L., M. Nowak, M. Quaddus, and M. Naude. 2014. Stakeholder engagement and sustainable corporate community investment. Business Strategy and the Environment 23 (7): 461–474.
Sandberg, J., C. Juravle, T.M. Hedesström, and I. Hamilton. 2009. The heterogeneity of socially responsible investment. Journal of Business Ethics 87 (4): 519–533.
Scholtens, B., and R. Sievänen. 2013. Drivers of socially responsible investing: A case study of four Nordic countries. Journal of Business Ethics 115 (3): 605–616.
Securities and Exchange Commission (SEC). 2010. Commission guidance regarding disclosure related to climate change. 17 CFR PARTS 211, 231 and 241 [Release Nos. 33-9106; 34-61469; FR-82]. https://www.sec.gov/rules/interp/2010/33-9106.pdf. Accessed 10 Jul 2021.
Talan, G., and G.D. Sharma. 2019. Doing well by doing good: A systematic review and research agenda for sustainable investment. Sustainability 11 (2): 353.
Tsang, S., R. Welford, and M. Brown. 2009. Reporting on community investment. Corporate Social Responsibility and Environmental Management 16 (3): 123–136.
UNEP Finance Initiative (UNEP FI). 2012. PSI Principles for Sustainable Insurance: A global sustainability framework and initiative of the United Nations Environment Programme Finance Initiative. https://www.unepfi.org/psi/wp-content/uploads/2012/06/PSI-document.pdf. Accessed 29 Jul 2019.
US SIF Foundation (USSIF). 2018. Report on US sustainable, responsible and impact investing trends 2018. Washington, DC: US SIF Foundation.
US SIF Foundation (USSIF). 2020. Report on US sustainable and impact investing trends 2020. Washington, DC: US SIF Foundation.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Gatzert, N., Reichel, P. Sustainable investing in the US and European insurance industry: a text mining analysis. Geneva Pap Risk Insur Issues Pract 49, 26–62 (2024). https://doi.org/10.1057/s41288-022-00270-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/s41288-022-00270-w