ホスピタリティ産業におけるESGパフォーマンスと財務成果：系統的文献レビュー

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:ESGと財務の相関を扱う際、測定手法の共線性と共通手法バイアスを統制する設計が不可欠だと示す。

🏢実務担当者:自己申告CSRの見栄えの良さと第三者評価の乖離を踏まえ、対外開示の検証可能性を高める必要がある。

🏛政策担当者:ESG評価機関の手法透明性と、ガバナンス開示の拡充を促す根拠になりうる。

Corporate social responsibility (CSR) and ESG are in the spotlight in hospitality management research; however, the CSR/ESG body of knowledge is fragmented by incompatible measurement approaches. Most of this work has been conducted at the regional level, with little to no attention to governance issues. To date, no study has combined a systematic review of the CSR/ESG literature in the hospitality industry with a pre‐registered design incorporating bibliometric analysis and a meta‐analytic approach. The main objectives are to map the research landscape on ESG/CSR to financial performance linkages for the hospitality industry (RQ1), to evaluate the combined strength of association and the direction of the ESG‐to‐performance linkage along with the structural moderators of this relationship through a meta‐analytic approach (RQ2), and to integrate the primary mediation processes, moderation variables, and research gaps, particularly about the governance dimension (RQ3). Peer‐reviewed empirical studies, along with non‐empirical studies, that have examined ESG or CSR as an independent variable and financial performance as a dependent variable for hospitality sector firms (hotels, restaurants, casino hotels, cruise lines, and resort complexes), limiting the language to English and the publication period to 2010 to 2026. We searched the Web of Science Core Collection and Scopus on January 12, 2026. We used six parallel search strings to address all three research questions, covering ESG/CSR, hospitality, and financial performance. Sensitivity was 100% because the search retrieved all 12 pre‐identified gold‐standard studies. Results showed a substantial duplication rate (49.6%) between the two databases. The protocol was registered with the Open Science Framework (OSF) before data collection (V5; all amendments were filed before each analytical stage). We assessed the empirical studies using a hybrid of the Newcastle‐Ottawa Scale (NOS, max 9) and a custom five‐item Econometric Quality Checklist. Risk‐of‐bias (RoB) inter‐rater reliability ( N  = 47 dual‐reviewed studies, 50% verification sample) reached Po = 83.5% and Cohen's κ  = 0.791 after resolving discrepancies, satisfying the pre‐registered ≥ 80% Po and ≥ 0.60 κ thresholds. All 47 RoB weight category assignments reached full agreement. Literature review studies were assessed using A Measurement Tool to Assess Systematic Reviews AMSTAR‐2. Ninety‐five studies included in total (72 empirical, 23 non‐empirical). K  = 23 studies ( k  = 84 effect sizes) were eligible for meta‐analytic pooling. The primary three‐level multilevel random‐effects model yielded r  = 0.211 (95% CI [0.103, 0.314], p  = 0.0001; I 2  = 93.3%). The one‐per‐study model yielded r  = 0.216 (95% CI [0.106, 0.320], p  = 0.0001). Restricting the pool to objective third‐party ESG ratings yielded r  = 0.061 ( p  = 0.195), while self‐reported CSR surveys yielded r  = 0.353 ( p  < 0.001). Funnel asymmetry was detected (Egger's test: z  = 2.732, p  = 0.013), but trim‐and‐fill imputed zero missing studies; PET‐PEESE bias‐corrected estimates are non‐significant, consistent with measurement‐type heterogeneity rather than suppression bias. The governance ESG dimension yielded r  = −0.023 from five studies. SciMat longitudinal co‐word analysis confirmed the ESG cluster growing from marginal (centrality = 8.37, P1) to structurally prominent (centrality = 17.71, P4), with the CSR‐to‐ESG terminological transition underway but still incomplete. The positive ESG‐to‐financial‐performance association in the hospitality industry is entirely dependent on measurement: objective third‐party ratings show no detectable association, whereas self‐reported property‐level CSR measures show a large positive effect. This divergence is a structurally diagnostic finding that generates two competing hypotheses: the construct validity hypothesis (enterprise‐level and property‐level ESG instruments target theoretically distinct constructs with different pathways to financial performance) and the methodological account (common method bias inflates self‐reported estimates), which cannot be disentangled because the ESG measurement approach and study design are fully collinear in the pool. The governance dimension is the field's primary intellectual deficit, established convergently across three evidence streams (bibliometric, thematic, researcher self‐report) and consistent with the meta‐analytic arm ( r  = −0.023, K  = 5), which is too sparse to estimate the relationship reliably and is itself a structural signal of the deficit. The primary value‐creation mechanisms, corporate reputation, competitive advantage, and ESG as organizational insurance during crises, are theoretically coherent but causally underidentified.

• semanticscholar https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/csr.71109first seen 2026-10-31 05:33:40 · last seen 2026-11-06 05:23:29