Esg Disclosure Firm Financial Performance Literature Review

Introduction

Environmental, social and governance (ESG) disclosure has moved from a peripheral reporting activity to a central concern of modern corporate finance. Where corporate reporting once concentrated almost exclusively on financial statements prepared for shareholders, listed firms now publish extensive non-financial information on carbon emissions, labour practices, board composition, supply-chain conduct and a widening range of sustainability metrics. This expansion has been driven by investor demand for material sustainability information, by the growth of professionally managed responsible-investment mandates, and by an increasingly prescriptive regulatory environment. The question that follows naturally from this shift, and the one that has generated one of the largest empirical literatures in finance and accounting, is whether disclosing more and better ESG information is associated with superior financial performance for the disclosing firm.

The question matters to several audiences at once. For managers, the answer bears directly on whether the costs of collecting, assuring and publishing sustainability data are recovered through tangible financial benefits or merely represent a compliance burden. For investors, it determines whether ESG information carries value-relevant signals that can be incorporated into asset pricing and portfolio construction. For regulators and standard-setters, the relationship speaks to whether mandating disclosure is likely to improve capital allocation or simply to raise reporting costs. Given these stakes, it is unsurprising that the academic response has been voluminous; yet the sheer volume of research has not produced a correspondingly clear consensus.

Instead, the empirical record is fragmented and, at first reading, contradictory. Some studies report that firms with stronger ESG disclosure enjoy higher accounting returns, higher market valuations and a lower cost of capital. Others find associations so small as to be economically trivial, and a number report null or even negative relationships once appropriate controls and identification strategies are applied. This divergence is not confined to isolated studies: large meta-analyses themselves reach different conclusions about the strength, and occasionally the direction, of the effect. The field is therefore characterised less by a settled answer than by a pattern of conflicting findings that resists simple summary.

Much of this apparent contradiction, this review argues, is traceable to how the two central constructs are defined and measured. ESG “disclosure” is operationalised variously as the quantity of information released, the quality or credibility of that information, or a third-party rating that conflates disclosure with underlying performance. “Financial performance”, likewise, is proxied by accounting measures, market-based valuation, stock returns and the cost of capital, which capture conceptually distinct outcomes. A central objective of this review is therefore not only to synthesise what the literature finds but to map how it measures the relationship, because measurement heterogeneity is itself a leading explanation for the conflicting results.

The purpose of this review is to synthesise the empirical and theoretical literature on the ESG disclosure–performance relationship, to make sense of why findings diverge, and to identify where the most important uncertainties remain. The review adopts a thematic structure. It first sets out the competing theoretical foundations that generate opposing predictions. It then examines the evidence supporting a positive relationship, before considering the conflicting, null and non-linear findings that qualify it. A fourth theme addresses measurement directly, mapping the constructs and proxies used across the literature and explaining how their heterogeneity drives inconsistency. The review then critically appraises the methodological features of the evidence base before setting out a defined research gap. For the purposes of this review, ESG disclosure is treated broadly, encompassing voluntary and mandatory non-financial reporting, and financial performance is treated as a family of related but distinct outcomes rather than a single construct.

Search Strategy

An illustrative structured search was conducted to identify peer-reviewed studies examining the relationship between ESG or corporate social responsibility (CSR) disclosure and corporate financial performance. Three electronic databases were searched, supplemented by backward and forward citation searching of key review articles and meta-analyses. The search combined terms relating to the disclosure construct (ESG disclosure, sustainability reporting, non-financial disclosure, CSR reporting), the performance construct (financial performance, firm value, cost of capital, Tobin’s Q, stock returns) and the firm-level unit of analysis. Boolean operators and truncation were used to broaden retrieval, and results were restricted to English-language sources in order to reflect the body of work most accessible to an international readership. The parameters applied are summarised in the table below.

The resulting set was deliberately weighted towards meta-analyses, large-sample empirical studies and influential narrative reviews, as these provide the most reliable basis for a thematic synthesis and allow the conflicting single studies they aggregate to be placed in context. The retained sources are not presented as an exhaustive mapping of a very large field but as a representative selection sufficient to illustrate the main currents of debate and, in particular, the measurement tensions that run through them. Because the review spans both the finance and accounting literatures, which have tended to emphasise different outcome variables, sources were purposively selected to represent accounting-based, market-based and cost-of-capital approaches to performance.

Themed Synthesis of the Literature

The retained literature was organised into four themes. The first establishes the competing theoretical frameworks that generate opposing predictions about the direction of any relationship. The second examines the substantial body of evidence consistent with a positive association. The third addresses the conflicting, null and non-linear findings that complicate that picture. The fourth, which is central to this review’s argument, examines how the relationship is measured and shows that construct and proxy heterogeneity is itself a principal source of the inconsistency documented in the first three themes.

Theoretical Foundations of the ESG Disclosure–Performance Link

The expectation that ESG disclosure should relate positively to financial performance rests on several complementary theoretical foundations. Stakeholder theory, most influentially articulated by Freeman (1984), holds that a firm’s long-run value depends on its ability to create and sustain relationships with a broad constituency of stakeholders, including employees, customers, suppliers, regulators and communities, and not shareholders alone. On this view, credible ESG disclosure signals that a firm is managing stakeholder relationships effectively, thereby reducing the risk of costly conflict, regulatory sanction or reputational damage and supporting durable cash flows. Legitimacy theory offers a related account, proposing that firms disclose social and environmental information to demonstrate that their conduct conforms to the norms of the society in which they operate, maintaining the implicit “licence to operate” on which continued profitability depends.

A distinct and arguably more precise mechanism is supplied by voluntary disclosure theory within financial economics. Drawing on the analysis synthesised by Verrecchia (2001), this perspective treats disclosure as a means of reducing information asymmetry between managers and outside investors. Where ESG factors are financially material but not captured by conventional financial statements, their disclosure provides investors with additional value-relevant information. Reduced information asymmetry narrows the premium that investors demand for bearing estimation risk, which should manifest as a lower cost of capital and, by extension, a higher firm valuation. Signalling theory complements this logic: high-quality firms may disclose extensively precisely to distinguish themselves from lower-quality peers that cannot credibly imitate such disclosure, so that disclosure functions as a costly and therefore informative signal.

Against these optimistic accounts stand theories that predict a neutral or negative relationship. The neoclassical or shareholder-primacy view, associated with the tradition that regards profit maximisation as the firm’s proper objective, treats discretionary expenditure on social and environmental activity, and the disclosure that accompanies it, as a diversion of resources that could otherwise be returned to shareholders. From this standpoint, extensive ESG disclosure may reflect managerial agency problems, in which executives pursue reputational or private benefits at shareholders’ expense. The trade-off hypothesis formalises this intuition by positing a genuine cost to sustainability investment that is not fully offset by financial return, implying a negative association. A further sceptical possibility is that disclosure is partly decoupled from substance: firms may engage in symbolic or selective reporting, sometimes characterised as greenwashing, in which case disclosure volume would carry little reliable information about either conduct or value.

These opposing predictions are important because they frame the empirical debate. If the positive-mechanism theories dominate, disclosure should be associated with lower risk and higher value; if the trade-off and agency accounts dominate, the association should be weak, negative or confined to firms where disclosure genuinely reflects material management. Crucially, the theories are not mutually exclusive, and several may operate simultaneously for different firms and in different institutional settings, which anticipates the heterogeneity that the empirical literature repeatedly encounters.

Evidence for a Positive Disclosure–Performance Relationship

A large and influential body of evidence is consistent with a positive relationship between ESG engagement, including its disclosure, and financial performance. The most frequently cited synthesis is the aggregated review by Friede, Busch and Bassen (2015), which pooled the findings of more than two thousand primary studies drawn from earlier reviews and vote-counting exercises. They reported that a large majority of studies found a non-negative relationship between ESG criteria and corporate financial performance, with positive findings substantially outnumbering negative ones. This headline result is widely interpreted as evidence that, on balance, attention to ESG factors is not financially detrimental and is frequently beneficial. Earlier meta-analytic work pointed in a broadly compatible direction: Orlitzky, Schmidt and Rynes (2003), synthesising several decades of research on corporate social and financial performance, reported a positive average association, albeit one whose magnitude varied with how both constructs were operationalised. The large meta-analysis by Margolis, Elfenbein and Walsh (2009) reached a compatible but more cautious conclusion, estimating a positive but small average relationship that was sensitive to specification and, in their reading, too modest to justify strong claims in either direction.

Beyond meta-analysis, several carefully designed single studies have strengthened the case for a causal, value-creating interpretation. Eccles, Ioannou and Serafeim (2014) compared matched samples of firms that had voluntarily adopted substantive sustainability policies and practices with those that had not, and reported that the high-sustainability firms subsequently exhibited superior accounting and stock-market performance over the long term, alongside distinctive governance and stakeholder-engagement processes. Flammer (2015) exploited a quasi-experimental setting, using closely contested shareholder proposals on CSR as a source of near-random variation, and found that the narrow passage of such proposals was associated with positive abnormal returns and subsequent operating improvements, an identification strategy that addresses the reverse-causality concern more convincingly than simple correlations.

A particularly coherent strand of evidence concerns the cost-of-capital channel predicted by voluntary disclosure theory. Dhaliwal et al. (2011) studied firms initiating standalone CSR reporting and reported that initiation was associated with a subsequent reduction in the cost of equity capital, particularly for firms with higher prior cost of capital, consistent with disclosure reducing information asymmetry. El Ghoul et al. (2011) found that firms with stronger CSR profiles exhibited a lower cost of equity, which they attributed to a broader investor base and reduced perceived risk. These findings are theoretically important because they identify a specific, economically interpretable mechanism linking disclosure to value, rather than relying on an unexplained correlation with accounting profit.

The materiality qualification advanced by Khan, Serafeim and Yoon (2016) refines this positive literature in an important way. Using a classification of sustainability issues by their financial materiality to particular industries, they reported that firms performing well on material sustainability issues outperformed those performing poorly on them, whereas performance on immaterial issues was unrelated, or even inversely related, to returns. This result reconciles part of the conflicting evidence by suggesting that the relationship is positive and economically meaningful only when disclosure and performance concern issues that are genuinely financially relevant to the firm, a nuance that undifferentiated aggregate measures necessarily obscure. Taken together, this theme provides substantial support for the proposition that credible, material ESG disclosure can be associated with lower risk and higher value, while already hinting that the effect is conditional rather than universal.

Conflicting, Null and Non-Linear Findings

Against the positive evidence stands a significant literature reporting weak, null, negative or non-linear relationships, and it is this body of work that prevents any confident general conclusion. The most fundamental challenge is one of causal identification. Much of the evidence is cross-sectional or, at best, based on panel correlations that cannot distinguish between competing explanations: that disclosure improves performance; that more profitable firms, possessing slack resources, can afford to invest in and disclose ESG activity; or that an unobserved factor, such as management quality or firm size, drives both. The slack-resources account, in which financial performance causes ESG engagement rather than the reverse, is theoretically well motivated and empirically difficult to exclude, and its plausibility means that positive correlations cannot be read straightforwardly as evidence that disclosure pays.

A second complication is that the direction and strength of reported associations vary systematically with context. The institutional environment appears to moderate the relationship: disclosure may carry more value-relevant information in settings where reporting is voluntary and therefore discriminating than in settings where it is mandated and near-universal. Industry matters because the financial materiality of environmental and social factors differs sharply across sectors, as the materiality literature implies. Country-level differences in legal protection, enforcement and investor preferences further condition the relationship, so that pooling heterogeneous samples can produce an average effect that describes no particular firm well. Narrative reviews of the broader corporate-finance literature, such as that by Gillan, Koch and Starks (2021), emphasise precisely this heterogeneity, noting that findings are highly sensitive to sample, period and specification and that the field has struggled to establish robust, generalisable causal effects.

A third possibility, increasingly discussed, is that the relationship is non-linear. Modest levels of ESG disclosure and engagement may be value-enhancing by reducing risk and improving stakeholder relations, whereas very high levels may reflect over-investment, or managerial empire-building in the guise of sustainability, that destroys value at the margin. Such a curvilinear, inverted-U pattern would appear as a positive, null or negative linear association depending on where in the distribution a particular sample happens to concentrate, offering a further explanation for apparently irreconcilable results. The distinction between disclosure and underlying performance is also critical here: firms may disclose extensively without corresponding substantive conduct, so that a measured association between disclosure volume and financial outcomes may capture reporting behaviour, selection, or greenwashing rather than any value-creating management of ESG risk.

Finally, the literature on mandatory disclosure regimes offers a cautionary counterpoint to the voluntary-disclosure findings. The analytical and empirical review by Christensen, Hail and Leuz (2021) concluded that although sustainability reporting can produce real effects, including on firm behaviour and capital markets, the evidence on net economic benefits is mixed and depends heavily on institutional context, enforcement and the comparability of the information produced. Their assessment underscores that the positive-disclosure mechanism documented in voluntary settings cannot be assumed to transfer automatically to mandated regimes, and that the costs of disclosure are real and may not always be recovered. This theme, therefore, does not simply contradict the positive literature; it specifies the conditions under which a positive relationship is and is not likely to hold, and it points firmly towards measurement and identification as the sources of the field’s inconsistency.

Measuring the Link: Constructs, Proxies and Rating Divergence

The central argument of this review is that much of the conflict surveyed above arises not from genuine disagreement about the world but from inconsistency in how the two constructs are measured. Both ESG disclosure and financial performance are latent constructs operationalised through a wide variety of proxies, and studies that reach opposing conclusions are frequently measuring different things under the same labels. Mapping this measurement heterogeneity is therefore essential to interpreting the literature coherently.

On the disclosure side, researchers have used at least three distinct families of measure. The first comprises disclosure-quantity scores, such as indices built from the presence or absence of items recommended by frameworks like the Global Reporting Initiative, or vendor “disclosure scores” that reward breadth of reporting irrespective of the underlying conduct. The second comprises third-party ESG ratings produced by commercial providers, which blend disclosure with assessments of policies, controversies and outcomes into a composite score. The third comprises researcher-constructed indices and content-analysis measures tailored to a specific study. These families are not interchangeable: a disclosure-quantity measure captures how much a firm reveals, whereas a rating purports to capture how well it performs, and conflating the two can produce spurious or contradictory findings. The distinction is not merely conceptual, because disclosure and underlying performance are themselves related: Clarkson et al. (2008), examining environmental reporting, found that firms with stronger environmental performance tended to disclose more voluntary environmental information, consistent with a signalling interpretation in which good performers reveal their type. Where this holds, disclosure volume proxies partly for performance; where it does not, as under greenwashing, the two diverge, and a measure that assumes their equivalence will mislead.

The problem is compounded by the now well-documented divergence among commercial ESG ratings. Berg, Kolbel and Rigobon (2022), analysing ratings from several major providers, reported that correlations between providers’ scores for the same firms were low, and they decomposed the divergence into differences in scope (which categories are measured), measurement (how a given category is assessed) and weighting (how categories are aggregated). The practical implication is severe: a study’s conclusion about whether ESG “pays” can depend substantially on which rating provider supplied the data, because the providers do not agree on what they are measuring. This “aggregate confusion” means that empirical findings built on different ratings are not directly comparable and that the literature’s apparent inconsistency is partly an artefact of inconsistent measurement at source.

On the performance side, an equivalent heterogeneity prevails. Studies use accounting-based measures such as return on assets and return on equity, which capture realised operating profitability but are backward-looking and subject to accounting discretion; market-based measures such as Tobin’s Q and the market-to-book ratio, which capture investors’ forward-looking valuation but incorporate expectations that may already price in ESG; stock-return measures and risk-adjusted abnormal returns, which test whether information is already impounded in prices; and cost-of-capital measures, which most directly test the disclosure mechanism but rely on contestable estimation. These proxies answer different questions, and a relationship that is positive on one may be absent on another. The following table maps the principal constructs, their common proxies and the type of question each addresses, drawing on representative studies from the themes above.

Read through this mapping, the conflicting literature becomes considerably more intelligible. A study that measures disclosure quantity against the cost of equity is testing the information-asymmetry mechanism and is likely to find a negative (beneficial) relationship; a study that measures a composite rating against contemporaneous stock returns is effectively asking whether the market has already priced ESG, and may well find nothing, because an efficient market would impound known information. Divergent ratings, mismatched proxies and differing time horizons can therefore generate opposing headline results from data that are, at a deeper level, consistent. The implication is that the question “does ESG disclosure improve financial performance?” is underspecified: it cannot be answered without stating which disclosure construct, which performance construct and which causal channel are intended. Measurement, in short, is not a technical footnote to this literature but one of its primary substantive findings.

Critical Appraisal of the Evidence

Taken together, the four themes reveal an evidence base that is extensive and sophisticated but constrained in ways that explain much of its apparent inconsistency. The most serious limitation is the difficulty of causal identification. The majority of reported associations rest on observational panel data in which disclosure, performance and numerous firm characteristics are jointly determined. Endogeneity arising from reverse causality, as in the slack-resources account, and from omitted variables such as management quality is pervasive, and only a minority of studies, such as Flammer (2015) with its regression-discontinuity design, employ identification strategies capable of supporting causal claims. Where strong designs are used, the estimated effects are often more modest and more conditional than the raw correlations suggest, which cautions against reading the large positive meta-analytic counts as direct evidence that disclosure causes value.

A second limitation concerns construct validity, the theme developed above. Because disclosure and performance are each measured in multiple non-equivalent ways, and because commercial ratings diverge substantially as Berg, Kolbel and Rigobon (2022) demonstrate, the literature aggregates findings that are not strictly comparable. Meta-analyses that pool across measures, even influential ones, risk combining heterogeneous estimands and reporting an average that corresponds to no well-defined causal quantity. The materiality finding of Khan, Serafeim and Yoon (2016) shows how much a measure’s validity matters: distinguishing material from immaterial issues changes not merely the magnitude but the economic interpretation of the relationship.

A third issue is selection and publication bias. Striking positive findings tend to be published and cited more readily than null results, and vote-counting syntheses that tally the direction of significant effects are particularly vulnerable to this bias. The dominance of positive counts in aggregate reviews may therefore partly reflect what reaches print rather than the underlying distribution of effects. Related to this is analytic flexibility: with many admissible measures of both constructs and many defensible specifications, researchers face a large garden of forking paths, and selective reporting of favourable combinations could inflate the apparent strength of the relationship.

A fourth limitation concerns generalisability. Much of the strongest evidence derives from large United States samples and from particular periods of rising investor attention to sustainability, and its transfer to other institutional settings, to emerging markets and to mandatory-disclosure regimes cannot be assumed, as Christensen, Hail and Leuz (2021) caution. Sample composition, the period studied and the regulatory backdrop all shape results, so that findings from one context may not hold in another. These limitations do not render the literature uninformative; rather, they indicate that its central inconsistencies are methodological and definitional, and that progress depends less on accumulating further correlations than on sharper identification and more disciplined measurement.

Identifying the Research Gap

The synthesis and appraisal above point to a coherent and well-defined gap. The early and most frequently asked questions, concerning whether ESG disclosure is associated with financial performance and whether more disclosure is better, have been substantially explored: associations exist, are often positive, but are conditional, modest under rigorous identification, and highly sensitive to how both constructs are measured. The pressing and under-addressed questions now concern causal mechanism, measurement validity and institutional contingency, rather than the mere existence of a correlation.

Three priorities would strengthen future research. First, the field needs more credible causal identification. Quasi-experimental designs exploiting regulatory changes, close votes, index-inclusion thresholds and other sources of exogenous variation offer a path beyond the observational correlations that dominate the literature, and they are essential if the policy question of whether mandated disclosure improves outcomes is to be answered. Second, research should confront measurement directly by disentangling disclosure quantity from disclosure quality and from underlying ESG performance, and by reporting results across multiple performance proxies so that the channel being tested is explicit. Given the documented divergence of commercial ratings, studies should either justify their choice of rating provider or demonstrate robustness across providers, and materiality-weighted measures should be preferred where the research question concerns financial relevance.

Third, the evidence base should be broadened beyond its concentration in United States voluntary-disclosure settings. The expansion of mandatory reporting regimes in Europe and elsewhere creates natural settings in which the costs and benefits of disclosure can be studied under changing rules, and emerging-market evidence is needed to test whether mechanisms established in developed markets generalise. Longer horizons would also help: if the value of disclosure operates through risk reduction and the gradual building of stakeholder relationships, short-window studies may understate effects that accrue over time. Addressing this gap would move the field beyond the unproductive and underspecified question of whether ESG disclosure “pays” towards the more useful goal of establishing for which firms, through which channel, under which disclosure regime and over which horizon a relationship holds.

Equally, future work would benefit from clearer conceptual separation of the outcome itself. Accounting profitability, market valuation and the cost of capital answer different questions, and treating them interchangeably has contributed to the appearance of conflict. Research that states in advance which outcome its theory predicts, and that aligns its disclosure measure with the proposed mechanism, would allow more precise hypotheses to be tested and would help to convert a large but fragmented body of correlations into cumulative, policy-relevant knowledge.

Conclusion

This review set out to synthesise the conflicting evidence on the relationship between ESG disclosure and firm financial performance and to locate the most important remaining uncertainties. Organised thematically, the literature shows a discernible structure beneath its apparent contradiction. Competing theories generate opposing predictions, with stakeholder, legitimacy and voluntary-disclosure accounts anticipating a positive relationship through reduced risk and information asymmetry, and shareholder-primacy and trade-off accounts anticipating a neutral or negative one. A substantial body of evidence, including the aggregated review by Friede, Busch and Bassen (2015), the quasi-experimental work of Flammer (2015) and the cost-of-capital findings of Dhaliwal et al. (2011) and El Ghoul et al. (2011), is consistent with a positive and economically interpretable relationship, particularly for material issues as Khan, Serafeim and Yoon (2016) show. Yet a considerable counter-literature documents weak, null, non-linear and context-dependent results, and much of this conflict can be traced to endogeneity and to inconsistent measurement.

The central message of the evidence is therefore one of conditionality and measurement sensitivity rather than uniform benefit or cost. The divergence of commercial ESG ratings documented by Berg, Kolbel and Rigobon (2022), and the multiplicity of non-equivalent performance proxies, mean that studies reaching opposing conclusions are often measuring different constructs under shared labels. The principal barriers to firmer conclusions are consequently methodological: weak causal identification, heterogeneous and divergent measurement, publication and specification bias, and limited generalisability beyond particular markets and periods. The clearest priority for future work is a programme of research that combines credible identification with disciplined, mechanism-aligned measurement across diverse institutional settings and longer horizons. Until such evidence accumulates, the most defensible reading is that credible, material ESG disclosure can be associated with lower risk and higher value for some firms through specific channels, but that confident general claims that disclosure uniformly improves financial performance are not supported by the evidence as it currently stands.

References

• Berg, F., Kolbel, J.F. and Rigobon, R. (2022) ‘Aggregate confusion: the divergence of ESG ratings’, Review of Finance, 26(6), pp. 1315–1344.
• Christensen, H.B., Hail, L. and Leuz, C. (2021) ‘Mandatory CSR and sustainability reporting: economic analysis and literature review’, Review of Accounting Studies, 26(3), pp. 1176–1248.
• Clarkson, P.M., Li, Y., Richardson, G.D. and Vasvari, F.P. (2008) ‘Revisiting the relation between environmental performance and environmental disclosure: an empirical analysis’, Accounting, Organizations and Society, 33(4–5), pp. 303–327.
• Dhaliwal, D.S., Li, O.Z., Tsang, A. and Yang, Y.G. (2011) ‘Voluntary nonfinancial disclosure and the cost of equity capital: the initiation of corporate social responsibility reporting’, The Accounting Review, 86(1), pp. 59–100.
• Eccles, R.G., Ioannou, I. and Serafeim, G. (2014) ‘The impact of corporate sustainability on organizational processes and performance’, Management Science, 60(11), pp. 2835–2857.
• El Ghoul, S., Guedhami, O., Kwok, C.C.Y. and Mishra, D.R. (2011) ‘Does corporate social responsibility affect the cost of capital?’, Journal of Banking & Finance, 35(9), pp. 2388–2406.
• Flammer, C. (2015) ‘Does corporate social responsibility lead to superior financial performance? A regression discontinuity approach’, Management Science, 61(11), pp. 2549–2568.
• Freeman, R.E. (1984) Strategic Management: A Stakeholder Approach. Boston: Pitman.
• Friede, G., Busch, T. and Bassen, A. (2015) ‘ESG and financial performance: aggregated evidence from more than 2000 empirical studies’, Journal of Sustainable Finance & Investment, 5(4), pp. 210–233.
• Gillan, S.L., Koch, A. and Starks, L.T. (2021) ‘Firms and social responsibility: a review of ESG and CSR research in corporate finance’, Journal of Corporate Finance, 66, 101889.
• Khan, M., Serafeim, G. and Yoon, A. (2016) ‘Corporate sustainability: first evidence on materiality’, The Accounting Review, 91(6), pp. 1697–1724.
• Margolis, J.D., Elfenbein, H.A. and Walsh, J.P. (2009) Does it pay to be good… and does it matter? A meta-analysis of the relationship between corporate social and financial performance. Working paper. Boston: Harvard Business School.
• Orlitzky, M., Schmidt, F.L. and Rynes, S.L. (2003) ‘Corporate social and financial performance: a meta-analysis’, Organization Studies, 24(3), pp. 403–441.
• Verrecchia, R.E. (2001) ‘Essays on disclosure’, Journal of Accounting and Economics, 32(1–3), pp. 97–180.